WO2017151638A1 - Magnetic resonance apparatus and program - Google Patents

Abstract

A magnetic resonance apparatus comprising an 8-channel receive coil apparatus; a channel identifying unit for identifying, of the eight channels, four channels (CH1, CH2, CH5, and CH6) disposed near an edge E1 of a liver, and four channels (CH3, CH4, CH7, and CH8) disposed at positions away from the edge E1 of the liver; and a respiratory signal generating unit for determining a first feature quantity of a navigator signal received by the four channels (CH1, CH2, CH5, and CH6), and a feature quantity of the navigator signal received by the four channels (CH3, CH4, CH7, and CH8), and generating a respiratory signal for a subject based on the first feature quantity and second feature quantity. A method provides a technique of generating a body motion signal, such as a respiratory signal or a heartbeat signal, for a subject while suppressing as much as possible the noise during performance of sequences.

Description

MAGNETIC RESONANCE APPARATUS AND PROGRAM

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] The present application claims priority to Japanese Patent Application No. 2016- 037834, filed on February 29, 2016, the entirety of which is incorporated herein by reference.

BACKGROUND

[0002] The present invention relates to a magnetic resonance apparatus for acquiring a body motion signal from a subject, and a program applied to the magnetic resonance apparatus.

[0003] A technique for body motion correction is known as a DC self-navigated method in the prior art. The DC self-navigated method acquires DC data representing data at the center of k-space. By using the DC data, it is possible to correct body motion. In the self-navigated method, however, an MR signal steeply diminishes until it reaches a steady state (see FIG. 51). Therefore, there is a problem that the signal value of a respiratory signal is not stabilized in a certain period of time D from the start of a scan.

[0004] Therefore, it is desired to provide a technique capable of stabilizing the signal value of a respiratory signal even before an MR signal reaches a steady state.

SUMMARY

[0005] A first aspect of the present invention is a magnetic resonance apparatus comprising a scanning section for performing a first scan for generating a first MR signal from a first body part including a moving body part of a subject, the first MR signal containing information on body motion of the subject; a coil having a plurality of channels for receiving the first MR signal generated by the first scan; a channel identifying unit for identifying n (n > 1) channels and m (m > 1) channels of the plurality of channels, the n channels being disposed near an edge of the moving body part, and the m channels being disposed at positions farther away from the edge of the moving body part than the n channels are; and a unit for determining a signal value of a body motion signal representing body motion of the subject, the unit determining a first feature quantity of the first MR signal received by the n channels and a second feature quantity of the first MR signal received by the m channels, and determining the signal value of the body motion signal based on the first feature quantity and the second feature quantity.

[0006] A second aspect of the present invention is a magnetic resonance apparatus comprising a scanning section for performing a first scan for generating a first MR signal from each of a plurality of slices defined in a first body part including a moving body part of a subject, the first MR signal containing information on body motion of the subject; a slice identifying unit for identifying u (u > 1) slices and v (v > 1) slices of the plurality of slices, the u slices being defined near an edge of the moving body part, and the v slices being defined at positions farther away from the edge of the moving body part than the u slices are; and a unit for determining a signal value of a body motion signal representing body motion of the subject, the unit determining a first feature quantity of the first MR signal obtained from the u slices and a second feature quantity of the first MR signal obtained from the v slices, and determining the signal value of the body motion signal based on the first feature quantity and the second feature quantity.

[0007] A third aspect of the present invention is a program applied to a magnetic resonance apparatus comprising a scanning section for performing a first scan for generating a first MR signal from a first body part including a moving body part of a subject, the first MR signal containing information on body motion of the subject, and a coil having a plurality of channels for receiving the first MR signal generated by the first scan, the program being for causing a computer to execute a channel identifying process of identifying n (n > 1) channels and m (m > 1) channels of the plurality of channels, the n channels being disposed near an edge of the moving body part, and the m channels being disposed at positions farther away from the edge of the moving body part than the n channels are; and a process of determining a first feature quantity of a first MR signal received by the n channels and a second feature quantity of the first MR signal received by the m channels, and determining a signal value of a body motion signal representing body motion of the subject based on the first feature quantity and the second feature quantity.

[0008] A fourth aspect of the present invention is a program applied to a magnetic resonance apparatus for performing a first scan for generating a first MR signal from each of a plurality of slices defined in a first body part including a moving body part of a subject, the first MR signal containing information on body motion of the subject, the program being for causing a computer to execute a slice identifying process of identifying u (u > 1) slices and v (v > 1) slices of the plurality of slices, the u slices being defined near an edge of the moving body part, and the v slices being defined at positions farther away from the edge of the moving body part than the u slices are; and a process of determining a first feature quantity of the first MR signal obtained from the u slices and a second feature quantity of the first MR signal obtained from the v slices, and determining a signal value of a body motion signal representing body motion of the subject based on the first feature quantity and the second feature quantity.

[0009] By using a first feature quantity of the first MR signal received by the n channels and a second feature quantity of the first MR signal received by the m channels, a body motion signal having a reduced impact of attenuation of the MR signal until the MR signal reaches a steady state may be determined. [0010] Moreover, by using a first feature quantity of the first MR signal obtained from the u slices and a second feature quantity of the first MR signal obtained from the v slices, a body motion signal having a reduced impact of attenuation of the MR signal until the MR signal reaches a steady state may be determined.

BRIEF DESCRIPTION OF THE DRAWINGS

[0011] FIG. 1 is a schematic diagram of a magnetic resonance apparatus in a first embodiment of the present invention.

[0012] FIG. 2 is a diagram illustrating the receive coil apparatus.

[0013] FIG. 3 is a diagram schematically showing a positional relationship between channels CHI to CH4 in the front array coil 4a and the liver.

[0014] FIG. 4 is a diagram schematically showing a positional relationship between channels CH5 to CH8 in the rear array coil 4b and the liver.

[0015] FIG. 5 is a diagram explaining units the processing apparatus implements.

[0016] FIG. 6 is a diagram explaining scans performed in a first embodiment.

[0017] FIG. 7 is a diagram explaining a sequence used in the prescan PS.

[0018] FIG. 8 is a diagram schematically showing a navigator signal 'a' and an imaging signal 'b' collected by performing sequences SAi to SA20.

[0019] FIG. 9 is a diagram showing a flow chart for performing the scans shown in FIG. 6. [0020] FIG. 10 is a diagram schematically showing slices defined at Step ST3 of FIG. 9. [0021] FIG. 11 is a diagram showing navigator data and imaging data collected by performing the sequences in a period of time Pi in the prescan PS.

[0022] FIG. 12 is a diagram showing navigator data and imaging data collected by performing the sequences in periods of time Pi to P_a in the prescan PS.

[0023] FIG. 13 is a diagram explaining a method of determining a signal value of a respiratory signal according to a method different from that in the first embodiment.

[0024] FIG. 14 is a diagram explaining a method of determining a signal value of a respiratory signal based on MR signals received at the channels CHI, CH2, CH5, and CH6.

[0025] FIG. 15 is a diagram schematically showing an integral value Si for combined data SYi.

[0026] FIG. 16 is a diagram schematically showing a respiratory signal.

[0027] FIG. 17 is a diagram showing two respiratory signals S_resi and S_res2.

[0028] FIG. 18 is a diagram showing a difference of the respiratory signals S_resi and S_res2 from a respiratory signal S_res3 derived by dividing the respiratory signal S_resi by the respiratory signal S_res2.

[0029] FIG. 19 is a diagram explaining a database.

[0030] FIG. 20 is a diagram explaining a scheme of obtaining a respiratory signal in a first embodiment. [0031] FIG. 21 is a diagram schematically showing a respiratory signal S_res4 obtained according to the method in the first embodiment.

[0032] FIG. 22 is an explanatory diagram for determining a signal value corresponding to a phase of respiration representing end expiration.

[0033] FIG. 23 is a diagram schematically showing a signal value rx corresponding to the phase of respiration representing end expiration.

[0034] FIG. 24 is a diagram explaining the main scan MS.

[0035] FIG. 25 is an explanatory diagram for performing the sequences SAi to SA20 in the main scan MS.

[0036] FIG. 26 is an explanatory diagram for determining a signal value of a respiratory signal.

[0037] FIG. 27 is an explanatory diagram for re-acquiring data in a period of time P₂.

[0038] FIG. 28 is a diagram showing a result of an experiment.

[0039] FIG. 29 is a diagram explaining a process the processing apparatus executes in a second embodiment.

[0040] FIG. 30 is a diagram explaining a sequence used in the prescan PS in the second embodiment.

[0041] FIG. 31 is a diagram showing MR signals A obtained by navigator sequences Ni to

Na. [0042] FIG. 32 is a diagram showing the flow of imaging in the second embodiment.

[0043] FIG. 33 is a diagram schematically showing a range of a body part to be imaged.

[0044] FIG. 34 is a diagram explaining the navigator sequence Ni.

[0045] FIG. 35 is a diagram explaining the navigator sequence N₂.

[0046] FIG. 36 is a diagram schematically showing a respiratory signal obtained by perforaiing the prescan PS.

[0047] FIG. 37 is a diagram explaining the main scan MS.

[0048] FIG. 38 is a diagram explaining a process the processing apparatus executes in a third embodiment.

[0049] FIG. 39 is a diagram schematically showing a body part to be imaged, and slices defined in the body part to be imaged in the third embodiment.

[0050] FIG. 40 is a diagram schematically showing a positional relationship between slices Xi to X₂o in axial planes and the liver.

[0051] FIG. 41 is a diagram explaining the prescan PS.

[0052] FIG. 42 is an explanatory diagram for performing sequences AXi to AX₂o in the prescan PS.

[0053] FIG. 43 is a diagram explaining a method of determining a signal value of a respiratory signal. [0054] FIG. 44 is a diagram schematically showing a respiratory signal Sresi i obtained according to the method in the third embodiment.

[0055] FIG. 45 is an explanatory diagram for determining a signal value corresponding to a phase of respiration representing end expiration.

[0056] FIG. 46 is a diagram schematically showing a signal value rx corresponding to the phase of respiration representing end expiration.

[0057] FIG. 47 is a diagram explaining the main scan MS.

[0058] FIG. 48 is an explanatory diagram for performing the sequences AXi to AX20 in the main scan MS.

[0059] FIG. 49 is an explanatory diagram for re-acquiring data in the period of time P₂.

[0060] FIG. 50 is a diagram showing an example for performing a channel identifying scan.

[0061] FIG. 51 is a waveform chart representing a temporal change of the signal intensity of an MR signal.

DETAILED DESCRIPTION

[0062] Now embodiments for practicing the invention will be described hereinbelow, although the present invention is not limited thereto.

[0063] In a first embodiment, FIG. 1 is a schematic diagram of a magnetic resonance apparatus in a first embodiment of the present invention. [0064] The magnetic resonance apparatus (referred to as "MR apparatus" hereinbelow) 1 comprises a magnet 2, a table 3, and a receive coil apparatus 4.

[0065] The magnet 2 has a reception space 21 in which a subject 13 is received. Moreover, the magnet 2 has coils, such as a superconductive coil 22, a gradient coil 23, and an RF coil 24. The superconductive coil 22 applies a static magnetic field, the gradient coil 23 applies a gradient pulse, and the RF coil 24 applies an RF pulse.

[0066] The table 3 has a cradle 3 a. The cradle 3 a is configured to be movable into the reception space 21. It is by the cradle 3a that the subject 13 is carried into the reception space 21. The receive coil apparatus 4 is attached to a torso of the subject 13.

[0067] FIG. 2 is a diagram explaining the receive coil apparatus 4. The receive coil apparatus 4 has a plurality of channels. While the following description will address a case in which the receive coil apparatus 4 has eight channels, the number of channels in the receive coil apparatus 4 is not limited to eight, and the present invention may be applied to cases in which the receive coil apparatus 4 has two or more channels.

[0068] The receive coil apparatus 4 has an anterior array coil 4a and a posterior array coil 4b. The anterior array coil 4a is a coil disposed on the front (abdominal) side of the subject 13, and has four channels CHI, CH2, CH3, and CH4. The four channels CHI to CH4 are arranged in a two-by-two array.

[0069] The posterior array coil 4b is a coil disposed on the rear (back) side of the subject 13, and has four channels CH5, CH6, CH7, and CH8. The four channels CH5 to CH8 are arranged in a two-by-two array. In the first embodiment, the anterior array coil 4a and posterior array coil 4b are attached near the liver.

[0070] FIG. 3 is a diagram schematically showing a positional relationship between the channels CHI to CH4 in the anterior array coil 4a and the liver. FIG. 3(a) shows the position of the channels in a zx-plane, and FIG. 3(b) shows the position of the channels in a d-d cross section of FIG. 3(a). In the first embodiment, an x-direction corresponds to a right-left (RL) direction, a y-direction corresponds to an anterior-posterior (AP) direction, and a z-direction corresponds to a superior-inferior (SI) direction.

[0071] The channels CHI and CH2 are arranged in the x-direction (RL-direction), and the channels CH3 and CH4 are also arranged in the x-direction (RL-direction). As compared with the channel CHI, the channel CH3 is the same in its position in the x-direction (RL-direction), but is different in its position in the z-direction (Si-direction). Similarly, as compared with the channel CH2, the channel CH4 is the same in its position in the x-direction (RL-direction), but is different in its position in the z-direction (Si-direction). The channels CHI and CH2 are disposed near an edge El of the liver adjacent to the lungs, while the channels CH3 and CH4 are disposed at positions farther away in a (-z)-direction from the edge El of the liver adjacent to the lungs than the channels CHI and CH2 are. For example, the channel CH3 is disposed near an edge E2 of the liver on a side opposite to the lung's side.

[0072] FIG. 4 is a diagram schematically showing a positional relationship between the channels CH5 to CH8 in the posterior array coil 4b and the liver. FIG. 4(a) shows the position of the channels in the zx-plane, and FIG. 4(b) shows the position of the channels in a d-d cross section of FIG. 4(a). [0073] The channels CH5 and CH6 are arranged in the x-direction, and the channels CH7 and CH8 are also arranged in the x-direction. As compared with the channel CH5, the channel CH7 is the same in its position in the x-direction, but is different in its position in the z-direction. Similarly, as compared with the channel CH6, the channel CH8 is the same in its position in the x-direction, but is different in its position in the z-direction. The channels CH5 and CH6 are disposed near the edge El of the liver, while the channels CH7 and CH8 are disposed at positions farther away in the (-z)-direction from the edge El of the liver adjacent to the lungs than the channels CH5 and CH6 are.

[0074] Referring back to FIG. 1, the description will be continued. The MR apparatus 1 further comprises a control section 5, a transmitter 6, a gradient power supply 7, a receiver 8, and a processing apparatus 9, a storage section 10, an operating section 11, and a display section 12.

[0075] The transmitter 6 supplies electric current to the RF coil 24, and the gradient power supply 7 supplies electric current to the gradient coil 23. The receiver 8 applies signal processing, such as demodulation/detection, to signals received from the receive coil apparatus 4. It should be noted that the magnet 2, control section 5, transmitter 6, and gradient power supply 7 together constitute the scanning section, and the receiver 8 constitutes the data generating unit.

[0076] The storage section 10 stores therein programs executed by the processing apparatus 9, and the like. It should be noted that the storage section 10 may be a non-transitory storage medium, such as a hard disk or a CD-ROM. The processing apparatus 9 loads a program stored in the storage section 10 and operates as a processor for executing processing written in the program. The processing apparatus 9 implements several kinds of unit by executing processing written in the programs. FIG. 5 is a diagram explaining the units the processor 9 implements. [0077] Channel identifying unit 91 identifies n channels and m channels from among the channels CHI to CH8 of the receive coil apparatus 4 based on a database, which will be discussed later (see FIG. 19), the n channels being disposed near the edge El of the liver (see FIGS. 3 and 4), and the m channels being disposed at positions farther away from the edge El of the liver than the n channels are. In the first embodiment, the n channels are four channels (CHI, CH2, CH5, and CH6), and the m channels are four channel (CH3, CH4, CH7, and CH8). Slice defining unit 92 defines slices based on information input from the operating section 11.

[0078] Respiratory signal generating unit 93 generates a respiratory signal, which will be discussed later. It should be noted that the respiratory signal generating unit 93 constitutes the unit for determining a signal value of a body motion signal.

[0079] Window defining unit 94 defines a window W, which will be discussed later (see FIG. 23), based on the respiratory signal.

[0080] Deciding unit 95 decides whether to accept imaging data as data for image

reconstruction or discard them each time a sequence is performed in a main scan MS, which will be discussed later.

[0081] The MR apparatus 1 comprises a computer including the processing apparatus 9. The processing apparatus 9 implements the channel identifying unit 91 to deciding unit 95, and the like by loading programs stored in the storage section 10. It should be noted that the processing apparatus 9 may implement the channel identifying unit 91 to deciding unit 95 by a single processor or by two or more processors. Moreover, the programs the processing apparatus 9 executes may be stored in a single storage section or separately in a plurality of storage sections. [0082] The operating section 11 is operated by an operator to input several kinds of information to the control section 5, processing apparatus 9, or the like. The display section 12 displays several kinds of information. The MR apparatus 1 is configured as described above.

[0083] FIG. 6 is a diagram explaining scans performed in the first embodiment. In the first embodiment, a localizer scan LS, a prescan PS, a main scan MS, etc. are performed.

[0084] The localizer scan LS is a scan for acquiring an image used in defining slices in the main scan MS, which will be discussed later.

[0085] The prescan PS is a scan for acquiring a respiratory signal required to define a window W, which will be discussed later (see FIG. 23).

[0086] The main scan MS is a scan for acquiring images in sagittal slices Ji to J20, which will be discussed later (see FIG. 10).

[0087] Now a sequence used in the prescan PS will be first described. FIG. 7 is a diagram explaining a sequence used in the prescan PS.

[0088] In FIG. 7, the prescan PS is illustrated separately in a plurality of periods of time Pi to P_a. In each period of time is performed a sequence set including sequences SAi to SA20 for collecting MR signals from the slices Ji to J20, which will be discussed later (see FIG. 10), according to a multi-slice method. Each period of time represents the repetition time TR. FIG. 7 shows a sequence set of the plurality of sequences SAi to SA20 performed in a period of time Pi of the periods of time Pi to P_a. FIG. 7 also schematically shows an example of the sequence SAi. The sequence SAi is configured to collect an MR signal (referred to as "navigator signal" below) 'a' containing information on body motion of the subject and an MR signal (referred to as "imaging signal" below) 'b' containing image information according to the DC self-navigated method.

[0089] The sequence SAi has an RF pulse a for exciting a slice, and a slice selective gradient pulse Gzl. It is by the RF pulse a and slice selective gradient pulse Gzl that a sagittal slice Ji is excited. Immediately after the slice selective gradient pulse Gzl, a rephaser pulse Gz2 is applied.

[0090] The sequence SAi has in a frequency encoding direction a dephaser pulse Gxl and a readout gradient pulse Gx2 for reading an MR signal. There is provided a wait time T_Wait between the rephaser pulse Gz2 after slice selection and the dephaser pulse Gxl. During the wait time T_Wait, a navigator signal 'a' for detecting subject's motion is collected. The navigator signal 'a' is a signal representing data (DC data) at the center of k-space. The wait time T_Wait is 20 μ8, for example. Moreover, an imaging signal 'b' in the sagittal slice Ji may be obtained by the readout gradient pulse Gx2.

[0091] The sequence SAi has in a phase encoding direction a phase encoding gradient pulse Gyl and a rephaser pulse Gy2 for refocusing the phase of spins. In FIG. 7, the magnetic field intensity of the phase encoding gradient pulse Gyl and the rephaser pulse Gy2 is denoted by "G."

[0092] After performing the sequence SAi, sequences SA₂ to SA₂o for acquiring images in the sagittal slices J₂ to J₂o are sequentially performed. The sequences SA₂ to SA₂o are represented by a similar sequence chart to that of the sequence SAi except the excitation frequency of the RF pulse a. Thus, in the period of time Pi, navigator signals 'a' and imaging signals 'b' in the sagittal slices Ji to J₂o are collected by performing the sequences SAi to SA₂o. FIG. 8 schematically shows the navigator signals 'a' and imaging signals 'b' collected by performing the sequences SAi to SA20. In FIG. 8, subscripts " 1," "2," "20" are added to symbol 'a' to distinguish the plurality of navigator signals 'a' obtained in the period of time Pi from one another. Likewise, subscripts " 1," "2," "20" are added to symbol 'b' to distinguish the imaging signals 'b' from one another.

[0093] After performing the sequences SAi to SA20 in the period of time Pi, the sequences SAi to SA20 are performed in a next period of time P₂ as well. Similarly thereafter, the sequences SAi to SA20 are repetitively performed. Therefore, the sequences SAi to SA20 are performed in each of the periods of time Pi to P_a.

[0094] Next, the flow for performing the scans (see FIG. 6) in the first embodiment will be described. FIG. 9 is a diagram showing a flow chart for performing the scans shown in FIG. 6.

[0095] At Step ST1, the operator puts the anterior array coil 4a and posterior array coil 4b (see FIG. 2) onto the subject. The operator estimates a rough position at which the subject's liver lies, and puts the anterior array coil 4a and posterior array coil 4b onto the subject so that the liver is positioned between them. The anterior array coil 4a (channels CHI, CH2, CH3, and CH4) are put onto the abdomen of the subject, while the posterior array coil 4b (channels CH5, CH6, CH7, and CH8) is put onto the back of the subject (see FIG. 2). After the receive coil apparatus 4 has been put onto the subject 13, the subject 13 is carried into the reception space 21 in the magnet 2, whereupon the flow goes to Step ST2.

[0096] At Step ST2, a localizer scan LS (see FIG. 6) is performed. The localizer scan LS is a scan for acquiring an image used for defining slices. After performing the localizer scan LS, the flow goes to Step ST3. [0097] At Step ST3, the operator operates the operating section 11 (see FIG. 1), and inputs information required to define slices in a main scan MS while referring to the image acquired by the localizer scan LS. Once the information has been input via the operating section 11, the slice defining unit 92 (see FIG. 5) defines slices based on the input information. FIG. 10

schematically shows the slices defined at Step ST3. The first embodiment shows an example for which twenty sagittal slices Ji to J20 are defined. After defining the sagittal slices Ji to J20, the flow goes to Step ST4.

[0098] At Step ST4, a prescan PS is performed. The prescan PS is a scan performed for generating a respiratory signal for the subject. Now the prescan PS will be described below.

[0099] FIGS. 11 and 12 are diagrams explaining the prescan PS. In performing the prescan PS, the control section 5 (see FIG. 1) sends data for an RF pulse in a sequence used in the prescan PS to the transmitter 6, and data for gradient pulses in the sequence used in the prescan PS to the gradient power supply 7. The transmitter 6 supplies electric current to the RF coil 24 based on the data received from the control section 5, while the gradient power supply 7 supplies electric current to the gradient coil 23 based on the data received from the control section 5. Thus, the prescan PS may be performed.

[0100] In the prescan PS, a sequence SAi is first performed in a period of time Pi. By performing the sequence SAi, a navigator signal ai and an imaging signal bi are collected from the sagittal slice Ji, as shown in FIG. 11. The navigator signal ai and imaging signal bi are received at the receive coil apparatus 4.

[0101] Since the receive coil apparatus 4 has the channels CHI to CH8 (see FIG. 2), the navigator signal ai and imaging signal bi are received at each of the channels CHI to CH8, and transmitted to the receiver 8 (see FIG. 1). The receiver 8 applies signal processing, such as demodulation/detection, to the signals received from each of the channels CHI to CH8, and outputs navigator data containing information (respiratory information) on the navigator signal ai and imaging data containing information (image information) on the imaging signal bi to the processing apparatus 9. Since in the first embodiment, the receive coil apparatus 4 has the eight channels CHI to CH8, the navigator data containing information (respiratory information) on the navigator signal ai and the imaging data containing information (image information) on the imaging signal bi may be obtained on a channel-by-channel basis. In FIG. 11, the navigator data obtained by the channels CHI to CH8 are respectively denoted by symbols An to A₁₈, while the imaging data obtained by the channels CHI to CH8 are respectively denoted by symbols Bn to Bis.

[0102] After performing the sequence SAi, and similarly thereafter, sequences SA₂ to SA₂o for acquiring data from the sagittal slices J₂ to J₂o are sequentially performed. By thus performing the sequences, navigator data and imaging data may be obtained on a channel-by- channel basis. FIG. 12 schematically shows the navigator data and imaging data obtained by performing the sequences SAi to SA₂o. In FIG. 12, navigator data and imaging data obtained by performing representative sequences SAi, SA₂, SAio, SAn, and SA₂o of the sequences SAi to SA₂o are shown for convenience of explanation. For example, navigator and imaging data obtained by performing the sequence SA₂ are denoted by symbols A₂i to A₂₈ and symbols B₂i to B₂₈, respectively.

[0103] It should be noted that by performing the sequences SAi to SA₂o, imaging data are acquired, as well as navigator data. However, the imaging data obtained in the prescan PS are not adopted as imaging data for image reconstruction, and are discarded. [0104] After performing the sequences SAi to SA20 in the period of time Pi, a signal value of a respiratory signal in the period of time Pi is determined. Now a technique of determining the signal value of a respiratory signal will be described below. In the following description, to clarify the effect of the present invention, an exemplary technique of determining the signal value of a respiratory signal according to a method different from that in the present embodiment will be described first before describing the technique of determining the signal value of a respiratory signal in the present embodiment.

[0105] FIG. 13 is a diagram explaining a method of determining the signal value of a respiratory signal according to a method different from that in the first embodiment.

[0106] FIG. 13(a) is a diagram schematically showing a positional relationship between the channels CHI to CH4 and the liver, while FIG. 13(b) is a diagram schematically showing a positional relationship between the channels CH5 to CH8 and the liver. The liver when the subject expires is indicated by a solid line, while the liver when the subject inspires is indicated by a dashed line.

[0107] First, let us consider the channels CHI to CH4 (see FIG. 13(a)). When the subject expires, the edge El of the liver adjacent to the lungs moves in the z-direction, so that the liver comes closer to the channels CHI and CH2. Therefore, it is considered that the signal values of received signals at the channels CHI and CH2 increase affected by the liver. On the other hand, it is considered that the signal values of received signals at the channels CH3 and CH4 decrease, as opposed to the channels CHI and CH2.

[0108] Next, consider a case in which the subject inspires. In this case, the edge El of the liver moves in the (-z)-direction, so that the liver goes away from the channels CHI and CH2. Therefore, it is considered that the signal values of received signals at the channels CHI and CH2 decrease. On the other hand, it is considered that the signal values of received signals at the channels CH3 and CH4 increase, as opposed to the channels CHI and CH2.

[0109] Therefore, received signals at the channels CHI and CH2 of the channels CHI to CH4 that lie near the edge El of the liver adjacent to the lungs increase when the subject expires and decrease when the subject inspires. On the other hand, received signals at the channels CH3 and CH4 that are disposed at positions away from the edge El of the liver adjacent to the lungs decrease when the subject expires and increase when the subject inspires. It can be seen from this that the times of the increase and decrease of the received signals at the channels CHI and CH2 are the reverse of those at CH3 and CH4.

[0110] Next, consider the channels CH5 to CH8 (see FIG. 13(b)). For the channels CH5 to CH8, the channels CH5 and CH6 of the channels CH5 to CH8 lie near the edge El of the liver. Therefore, received signals at the channels CH5 and CH6 increase when the subject expires and decrease when the subject inspires, as with the received signals at the channels CHI and CH2 (see FIG. 13(a)). On the other hand, received signals at the channels CH7 and CH8 decrease when the subject expires and increase when the subject inspires, as with those received at the channels CH3 and CH4 (see FIG. 13(a)).

[0111] Therefore, as shown in FIG. 13(c), the times of the increase and decrease of the received signals at the channels CHI, CH2, CH5, and CH6 are the reverse of those at the channels CH3, CH4, CH7, and CH8. Accordingly, when all the received signals at the eight channels CHI to CH8 are added together, the amplitude of the respiratory signal cannot be increased, resulting in a problem that it is difficult to obtain a respiratory signal fully reflecting subject's respiratory motion.

[0112] Then, as an example, a method of generating a respiratory signal may be

contemplated to involve selecting channels at which received signals increase/decrease at the same timing from among the channels CHI to CH8, and generating a respiratory signal based on signals received at the selected channels. As described above, those of the channels CHI to CH8 that lie near the edge El of the liver tend to have signals increasing/decreasing approximately at the same timing. Accordingly, it may be contemplated to select channels CHI, CH2, CH5, and CH6 lying near the edge El of the liver from among the channels CHI to CH8, and determining the signal value based on MR signals received by these channels. Now the method of determining the signal value of a respiratory signal based on MR signals received at the channels CHI, CH2, CH5, and CH6 will be described below.

[0113] FIG. 14 is a diagram explaining the method of determining the signal value of a respiratory signal based on MR signals received at the channels CHI, CH2, CH5, and CH6.

[0114] First, navigator data (those hatched in FIG. 14) obtained from the channels CHI, CH2, CH5, and CH6 are added together to thereby combine navigator data together, thus determining combined data SYi.

[0115] After acquiring the combined data SYi, time-integration is applied to the combined data SYi to determine an integral value Si. FIG. 15 schematically shows an integral value Si for the combined data SYi. The integral value Si thus determined is used as a signal value of the respiratory signal in the period of time Pi. [0116] Similarly thereafter, the sequences SAi to SA20 are performed in each of the periods of time P₂ to P_a, and integral values for combined data are calculated. Thus, a respiratory signal in performing the prescan PS may be determined (see FIG. 16).

[0117] FIG. 16 is a diagram schematically showing a respiratory signal. Only navigator data obtained by the channels CHI, CH2, CH5, and CH6 are combined (added) together, whereby a respiratory signal S_resi having its signal value varying according to subject's respiration may be acquired.

[0118] It can be seen that the respiratory signal S_resi reflects a subject's respiratory motion because it has the signal value increasing/decreasing with time. In the self-navigated method, however, an MR signal steeply diminishes until it reaches a steady state (see FIG. 51).

Therefore, there is a problem that the signal value of the respiratory signal S_resi is not stabilized in a certain period of time D from the start of the prescan PS.

[0119] Accordingly, the inventor of the present application has made an intensive study and come up with a technique for solving the problem. Now a principle of solving the problem will be briefly described below referring to FIGS. 17 and 18.

[0120] In FIG. 17, two respiratory signals S_resi and S_res2 are schematically shown. The respiratory signal S_resi schematically represents a respiratory signal obtained based on MR signals received at the channels CHI, CH2, CH5, and CH6, while the respiratory signal S_res2 schematically represents a respiratory signal obtained based on MR signals received at the channels CH3, CH4, CH7, and CH8. The respiratory signals S_resi and S_res2 both have instable signal values in the period of time D. The inventor of the present application, however, found that the impact of attenuation of an MR signal until it reaches a steady state may be reduced by dividing the respiratory signal S_resi by the respiratory signal S_res2. FIG. 18 schematically shows a difference of the respiratory signals S_resi and S_res2 from a respiratory signal S_res3 derived by dividing the respiratory signal S_resi by the respiratory signal S_res2. By dividing the respiratory signal S_resi by the respiratory signal S_res2, the respiratory signal may be stabilized in the period of time D from the start of the prescan PS and until the MR signal reaches a steady state.

Moreover, the respiratory signal S_resi has times of increase and decrease offset from the respiratory signal S_res2 by about half a period. Therefore, the amplitude of the respiratory signal Sres3 may also be increased by dividing the respiratory signal S_resi by the respiratory signal S_res2. It is thus possible to obtain a high-quality respiratory signal from immediately after the start of the prescan.

[0121] In the present embodiment, a respiratory signal is determined based on the principle described referring to FIG. 18. Now a scheme of determining a respiratory signal in the first embodiment will be described below.

[0122] With regard to determining a respiratory signal in the first embodiment, the channel identifying unit 91 (see FIG. 5) refers to a database stored in the storage section 10 before performing the prescan PS. FIG. 19 is a diagram explaining the database. In the database, information on the channels in the receive coil apparatus 4 is registered. The information on the channels in the receive coil apparatus 4 is registered in the database beforehand before scanning the subject. Now the database will be described below.

[0123] In the database, a field 'a' representing a receive coil apparatus, a field 'b' representing a channel in the receive coil apparatus, and a field 'c' representing the position of the channel relative to the edge El of the liver are registered. A hollow circle in the field 'c' indicates that the channel is disposed near the edge El of the liver, and a hollow triangle in the field 'c' indicates that the channel is disposed at a position away from the edge El of the liver. In the first embodiment, n channels of the eight channels CHI to CH8 are registered as the channel disposed near the edge El of the liver, and m channels thereof are registered as the channel disposed at a position away from the edge El of the liver. Here, n=4, that is, four channels CHI, CH2, CH5, and CH6 are registered as the channel disposed near the edge El of the liver, and m=4, that is, four channels CH3, CH4, CH7, and CH8 are registered as the channel disposed at a position away from the edge El of the liver.

[0124] The channel identifying unit 91 refers to the database (see FIG. 19) to identify the channels CHI, CH2, CH5, and CH6 registered as the channel disposed near the edge El of the liver, and the channels CH3, CH4, CH7, and CH8 registered as the channel disposed at a position away from the edge El of the liver based on the information in the field 'c' in the database.

[0125] After identifying the channels, the prescan PS is performed to obtain navigator data, as shown in FIG. 12. Then, a signal value of a respiratory signal is determined. Specifically, the signal value of the respiratory signal is determined as follows.

[0126] FIG. 20 is a diagram explaining a scheme of determining a respiratory signal in the first embodiment. The respiratory signal generating unit 93 (see FIG. 5) combines together navigator data (those hatched in FIG. 20) obtained by the channels CHI, CH2, CH5, and CH6. Here, the respiratory signal generating unit 93 combines these navigator data together by adding them together. This gives combined data SYi.

[0127] After acquiring the combined data SYi, the respiratory signal generating unit 93 determines a feature quantity of the navigator signals received by the channels CHI, CH2, CH5, and CH6 based on the combined data SYi. In the first embodiment, the combined data SYi is time-integrated to thereby calculate an integral value (area) Si, which is determined as the feature quantity of the navigator signals received by the channels CHI, CH2, CH5, and CH6.

[0128] Next, the respiratory signal generating unit 93 combines (adds) together navigator data (those non-hatched in FIG. 20) obtained by the channels CH3, CH4, CH7, and CH8. This gives combined data SY₂.

[0129] After acquiring the combined data SY₂, the respiratory signal generating unit 93 determines a feature quantity of the navigator signals received by the channels CH3, CH4, CH7, and CH8 based on the combined data SY₂. In the first embodiment, the combined data SY₂ is time-integrated to thereby calculate an integral value (area) S₂, which is determined as the feature quantity of the navigator signals received by the channels CH3, CH4, CH7, and CH8.

[0130] Next, the respiratory signal generating unit 93 determines a ratio r between the integral values Si and S₂. In FIG. 20, this is represented as r=Si/S₂=ri. In the present embodiment, the ratio r between the integral values is adopted as the signal value r of the respiratory signal for the subject.

[0131] Similarly thereafter, the sequences SAi to SA₂o are performed in each of the periods of time P₂ to P_a to calculate the ratio r between the integral values of the combined data. Thus, signal values r=n, r₂, r₃, r_a of the respiratory signal in the periods of time Pi, P₂, P₃, P_a may be determined (see FIG. 21).

[0132] FIG. 21 is a diagram schematically showing a respiratory signal S_res4 determined according to the method in the first embodiment. The respiratory signal generating unit 93 determines the integral value Si for the combined data SYi and the integral value S₂ for the combined data SY₂, and determines the ratio r between the integral values as a signal value of the respiratory signal. By thus calculating a ratio r between integral values, the impact of steep attenuation of an MR signal encountered in a period of time from immediately after the start of the prescan PS and until the MR signal reaches a steady state may be reduced. Therefore, a stable respiratory signal may be generated from immediately after the start of the prescan PS. Moreover, by calculating the ratio r between integral values, the amplitude of the respiratory signal may be increased.

[0133] While the sequences used in the prescan PS have gradient pulses Gyl and Gy2 (see FIG. 7) in the phase encoding direction, the prescan PS is not a scan performed for determining images in the slices Ji to J₂o. Therefore, the values of the magnetic field intensity G of the gradient pulses Gyl and Gy2 in the phase encoding direction may be unchanged during the prescan PS. Typically the prescan PS may be performed with the magnetic field intensity G set to G=0.

[0134] After determining the respiratory signal S_res4, the flow goes to Step ST5. At Step ST5, based on the respiratory signal S_res4, a window is defined for deciding whether to accept imaging data acquired in a main scan MS (Step ST6), which will be discussed later, as data for image reconstruction or discard them.

[0135] Now a method of defining the window will be described. First, the window defining unit 94 determines a signal value corresponding to a phase of respiration representing end expiration based on the signal value of the respiratory signal S_res4. [0136] FIG. 22 is an explanatory diagram for determining a signal value corresponding to the phase of respiration representing end expiration. In FIG. 22, a positional relationship between the channels CHI to CH8 and the liver is schematically shown. The liver when the subject expires is indicated by a solid line, while the liver when the subject inspires is indicated by a dashed line.

[0137] Let us first consider a case in which the subject expires. When the subject expires, the edge El of the liver moves in the S-direction, so that the liver comes closer to the channels CHI, CH2, CH5, and CH6. Therefore, the signal values of received signals at the channels CHI, CH2, CH5, and CH6 increase affected by the liver, resulting in a large integral value Si for the combined data obtained by the channels CHI, CH2, CH5, and CH6. On the other hand, the signal values of received signals at the channels CH3, CH4, CH7, and CH8 decrease, resulting in a small integral value S₂ for the combined data obtained by the channels CH3, CH4, CH7, and CH8. Therefore, when the subject expires, the ratio r between integral values has a large value.

[0138] Next, consider a case in which the subject inspires. When the subject inspires, the edge El of the liver moves in the I-direction. Therefore, signal values of received signals at the channels CHI, CH2, CH5, and CH6 decrease, resulting in a small integral value Si for the combined data obtained by the channels CHI, CH2, CH5, and CH6. On the other hand, the integral value S₂ for the combined data obtained by the channels CH3, CH4, CH7, and CH8 has a large value. Therefore, when the subject inspires, the ratio r between integral values has a small value.

[0139] As such, r has a large value when the subject expires, while it has a small value when the subject inspires. Accordingly, in the first embodiment, a local maximum (which is a signal value when the signal value of the respiratory signal changes from an increase to a decrease) of the respiratory signal is decided to be a signal value corresponding to a phase of respiration representing end expiration. FIG. 23 schematically shows a signal value rx corresponding to the phase of respiration representing end expiration determined by the window defining unit 94.

[0140] After determining the signal value rx corresponding to the phase of respiration representing end expiration, the window defining unit 94 defines a window W based on the signal value rx. For example, the window W is defined as follows.

[0141] The window defining unit 94 first determines a difference AD between the maximum and minimum of the respiratory signal. A range W is then defined over y % (for example, y=20) of the difference AD centered on the signal value rx. The thus-defined range W is determined as the window W for deciding whether or not to accept imaging data as data for image

reconstruction.

[0142] After defining the window W, the flow goes to Step ST6. At Step ST6, a main scan MS for acquiring images in the sagittal slices Ji to J20 (see FIG. 10) is performed.

[0143] FIG. 24 is a diagram explaining the main scan MS. The main scan MS performs sequences SAi to SA20 in periods of time Pi to Pb. The sequences SAi to SA20 performed in the main scan MS are sequences according to the DC self-navigated method, as with those performed in the prescan PS. Therefore, in the main scan MS, again, navigator signals (ai to a₂o) and imaging signals (bi to b₂o) are collected by performing the sequences SAi to SA20. Now the main scan MS will be particularly described below. [0144] FIG. 25 is an explanatory diagram for performing the sequences SAi to SA20 in the main scan MS. First, a sequence SAi is performed in a period of time Pi. By performing the sequence SAi, a navigator signal ai and an imaging signal bi are collected. The magnetic field intensity G of the gradient pulses in the phase encoding direction for the sequence SAi is set to a value G=0 for obtaining imaging data on a line at ky=0. The navigator signal ai and imaging signal bi are received at the receive coil apparatus 4.

[0145] The navigator signal ai is received at each of the channels CHI to CH8 in the receive coil apparatus 4, and transmitted to the receiver 8. The receiver 8 applies signal processing, such as demodulation/detection, to the signals received from the receive coil apparatus 4, and outputs navigator data An to Ais containing information (respiratory information) on the navigator signal ai and imaging data Bn to Bis containing information (image information) on the imaging signal bi to the processing apparatus 9.

[0146] After performing the sequence SAi, and similarly thereafter, sequences SA₂ to SA20 for acquiring images in the sagittal slices J₂ to J20 are sequentially performed. The magnetic field intensity G of the gradient pulses in the phase encoding direction for the sequences SA₂ to SA20 is set to a value G=0 for obtaining imaging data on a line at ky=0. Therefore, imaging data on a line at ky=0 is acquired from each of the sagittal planes Ji to J20 in the period of time Pi.

[0147] After performing the sequences SAi to SA20 in the period of time Pi, signal values of a respiratory signal in the period of time Pi are determined as follows.

[0148] FIG. 26 is an explanatory diagram for determining the signal value of a respiratory signal. The respiratory signal generating unit 93 combines navigator data obtained from the channels CHI, CH2, CH5, and CH6 together to generate combined data SYi. After generating the combined data SYi, the respiratory signal generating unit 93 time-integrates the combined data SYi to calculate an integral value Si.

[0149] Next, the respiratory signal generating unit 93 combines the navigator data obtained from the channels CH3, CH4, CH7, and CH8 together to generate combined data SY₂. After generating the combined data SY₂, the respiratory signal generating unit 93 time-integrates the combined data SY₂ to calculate an integral value S₂.

[0150] After calculating these integral values Si and S₂, the respiratory signal generating unit 93 calculates a ratio r between the integral values Si and S₂. In FIG. 26, this is represented as r=Si/S₂=ri. Therefore, the ratio r=n between the integral values is adopted as the signal value r of the respiratory signal for the subject in the period of time Pi.

[0151] After determining the ratio r between integral values, the deciding unit 95 (see FIG. 5) decides whether or not to use imaging data Bn to B₂os collected in the period of time Pi as data for image reconstruction based on the ratio r=n between integral values for the respiratory signal. In FIG. 26, the signal value (ratio between integral values) r=n in the period of time Pi does not fall within the window W, and therefore, the imaging data Bn to B₂os collected in the period of time Pi are discarded. After performing the sequences in the period of time Pi, the process moves to a period of time P₂.

[0152] In the period of time P₂, data discarded in the period of time Pi are re-acquired. FIG. 27 is an explanatory diagram for re-acquiring data in the period of time P₂.

[0153] In the period of time P₂, the magnetic field intensity G of the gradient pulses in the phase encoding direction for the sequences is set to a value G=0 for obtaining imaging data on a line at ky=0, as in the period of time Pi. Therefore, in the period of time P₂, imaging data on a line at ky=0 are obtained from each of the sagittal slices Ji to ho.

[0154] After performing the sequences SAi to SA₂o in the period of time P₂, the respiratory signal generating unit 93 generates combined data SYi and SY₂, and calculates an integral value Si for the combined data SYi and an integral value S₂ for the combined data SY₂. The respiratory signal generating unit 93 then calculates a ratio r between the integral values. In FIG. 27, the ratio r in the period of time P₂ is represented as r=Si/S₂=r₂. Therefore, r=r₂ is used as the signal value of the respiratory signal in the period of time P₂.

[0155] Next, the deciding unit 95 decides whether or not the signal value of the respiratory signal falls within the window W. Referring to FIG. 27, the signal value (ratio between integral values) r=r₂ in the period of time P₂ falls within the window W. Therefore, the deciding unit 95 decides that the imaging data Bn to B₂os collected in the period of time P₂ are accepted as data for image reconstruction.

[0156] Similarly thereafter, the sequences SAi to SA₂o are performed, and whether or not the signal value of the respiratory signal in the main scan MS falls within the window W is decided in each period of time. In the case that the signal value does not fall within the window W, the sequences SAi to SA₂o are performed in a next period of time without changing the value of the magnetic field intensity G of the gradient pulses in the phase encoding direction. On the other hand, in the case that the signal value falls within the window W, the value of the magnetic field intensity G for the gradient pulses in the phase encoding direction is changed, and the sequences SAi to SA₂o for acquiring imaging data on another line in ky is performed in the next period of time. The sequences SAi to SA₂o are repetitively performed until all data in k-space required in image reconstruction are acquired. Once all data in k-space required in image reconstruction have been acquired, images in the sagittal slices Ji to J20 (see FIG. 10) are reconstructed, and the flow of the first embodiment is terminated.

[0157] In the first embodiment, combined data SYi is determined based on navigator data obtained by the channels CHI, CH2, CH5, and CH6, and combined data SY2 is determined based on navigator data obtained by the channels CH3, CH4, CH7, and CH8. Then, an integral value S i for the combined data SYi and an integral value S2 for the combined data SY2 are determined, and a ratio r between the integral values is determined as a signal value of the respiratory signal. By thus calculating the ratio r between integral values, the impact of steep attenuation of an MR signal encountered in the period of time D (which is a period of time from immediately after the start of the prescan PS and until the MR signal reaches a steady state) may be reduced. Therefore, a stable respiratory signal may be generated from immediately after the start of the prescan PS. Moreover, by calculating the ratio r between integral values, the amplitude of the respiratory signal may be increased.

[0158] In the first embodiment, after determining the respiratory signal S_res4 by the prescan PS, a window W is defined (see FIG. 23). A main scan MS is then performed, and in the case that the signal value of the respiratory signal does not fall within the window W, imaging data are discarded, so that only imaging data collected when the signal value of the respiratory signal falls within the window W are accepted. Therefore, an image having reduced body motion artifacts may be obtained. [0159] As described above, in the first embodiment, a high-quality respiratory signal may be obtained by calculating the ratio r between integral values. To verify this, an experiment was performed using a phantom. Now results of the experiment will be described below.

[0160] FIG. 28 is a diagram showing results of the experiment. FIG. 28 shows three respiratory signals S_res5, S_res6, and S_reS7. The respiratory signal S_res5 (comparative example 1) represents a respiratory signal determined by calculating an integral value for combined data of navigator data from the channels CHI, CH2, CH5, and CH6. The respiratory signal S_res6 (comparative example 2) represents a respiratory signal determined by calculating an integral value for combined data of navigator data from the channels CH3, CH4, CH7, and CH8. The respiratory signal S_reS7 represents a respiratory signal determined according to the method in the first embodiment (the method of calculating a ratio r between integral values).

[0161] It can be seen that the respiratory signals S_res5 and S_res6 both have instable signal values of the respiratory signals in the period of time D affected by attenuation of the MR signal. On the other hand, it can be seen that the respiratory signal S_reS7 has stable signal values of the respiratory signal in the period of time D because the impact of attenuation of the MR signal is reduced in the period of time D.

[0162] It should be noted that in the first embodiment, a phase of respiration when the subject completely expires is identified to define a window W based on the phase of respiration. However, it is possible to identify another phase of respiration (for example, a phase of respiration when the subject completely inspires, or a phase of respiration in the middle of subject's inspiration or expiration) different from that when the subject completely expires, and define the window W based on the different phase of respiration. [0163] Moreover, in the first embodiment, n=4, that is, four channels CHI, CH2, CH5, and CH6 are registered in the database (see FIG. 19) as the n channels of the eight channels CHI to CH8 that are disposed near the edge El of the liver. However, it is not necessary to register all of the four channels CHI, CH2, CH5, and CH6, and a single (i.e., n=l), two (i.e., n=2), or three (i.e., n=3) of the four channels CHI, CH2, CH5, and CH6 may be registered therein insofar as a high-quality respiratory signal can be generated. Moreover, m=4, that is, four channels CH3, CH4, CH7, and CH8 are registered in the database (see FIG. 19) as the m channels of the eight channels CHI to CH8 that are disposed at positions away from the edge El of the liver.

However, it is not necessary to register all of the four channels CH3, CH4, CH7, and CH8, and a single (i.e., m=l), two (i.e., m=2), or three (i.e., m=3) of the four channels CH3, CH4, CH7, and CH8 may be registered therein insofar as a high-quality respiratory signal can be generated.

[0164] In the first embodiment, it is presupposed that when the receive coil apparatus 4 is put onto the subject, the channels CHI, CH2, CH5, and CH6 are positioned near the edge El of the liver. Therefore, the channels CHI, CH2, CH5, and CH6 are registered in the database as the channel lying near the edge El of the liver, while the channels CH3, CH4, CH7, and CH8 are registered as the channel disposed at a position away from the edge El of the liver. However, the size of each channel, the orientation of channel arrangement, the positional relationship between the channels and the liver, the number of channels included in the coil, etc. vary according to the type of the coil. Therefore, when generating a respiratory signal using a coil apparatus different from the receive coil apparatus 4, channels of a plurality of channels in the different coil apparatus that are disposed near the edge El of the liver and channels disposed at positions away from the edge El of the liver may be registered in the database. [0165] In the first embodiment, the slice defining unit 92 defines slices based on the information input by the operator via the operating section 11. However, the slice defining unit 92 may analyze an image obtained by the localizer scan LS to automatically define slices.

[0166] In the first embodiment, an integral value for combined data is calculated as a feature quantity for MR signals collected by channels. However, another feature quantity may be determined in place of the integral value insofar as a respiratory signal can be generated. For example, a maximum of combined data may be determined as the feature quantity. In the case that the maximum of combined data is determined, a ratio of maxima of the combined data is used as a signal value of the respiratory signal.

[0167] In the first embodiment, the order of data collection is defined in the main scan MS so that imaging data on a line at ky=0 are first collected. However, the present invention is not limited to the order of data collection, and it may be applied to the case in which the imaging data are collected in any order of data collection (for example, in a sequential order or in a centric order).

[0168] While the first embodiment addresses a case in which the respiratory signal is generated using a DC self-navigated sequence, a case in which a different navigator sequence is used to generate a respiratory signal will be described in a second embodiment. It should be noted that the hardware configuration of the MR apparatus is the same as that in the first embodiment.

[0169] FIG. 29 is a diagram explaining processing the processing apparatus 9 executes in the second embodiment. As compared with the first embodiment, the second embodiment has range defining unit 921, in place of the slice defining unit 92, for defining a range of the imaged body part, and the other units are the same as those in the first embodiment.

[0170] The processing apparatus 9 in the second embodiment is an example for configuring the channel identifying unit 91 to deciding unit 95, and it functions as these units by loading the programs stored in the storage section 10.

[0171] FIG. 30 is a diagram explaining a sequence used in a prescan PS in the second embodiment. In the prescan PS, a plurality of navigator sequences Ni to N_a are performed. The navigator sequences will be described hereinbelow. Since the navigator sequences Ni to N_a are expressed by the same sequence chart, the following description of the navigator sequences will focus upon a representative navigator sequence Ni from among the navigator sequences Ni to

[0172] The navigator sequence Ni is a sequence performed for acquiring a respiratory signal for the subject. The navigator sequence Ni has an excitation pulse EXi and a killer pulse K. The excitation pulse EXi is applied by the RF coil 24, and the killer pulse K is applied by the gradient coil 23. In the first embodiment, no gradient pulse is applied while the excitation pulse EXi is being applied. Thus, since the excitation pulse EXi is a non-selective RF pulse for exciting the subject without slice selection, a wide range of a body part (torso including the liver and lungs, for example) may be excited by applying the excitation pulse EXi. Since no gradient pulse is applied during excitation in the first embodiment, the excitation may be achieved without making loud noise. After applying the excitation pulse EXi, an MR signal A representing data at the center of k-space is collected in a data collection period DA. After the data collection period DA, the killer pulse K for canceling transverse magnetization is applied. The killer pulse K may be applied in any one of the axes Gx, Gy, Gz. The first embodiment shows a case in which it is applied in the Gx axis. It should be noted that smaller slew rates SR for a rise time Tu and a fall time Td of the killer pulse K is more desirable for reducing noise during performance of the navigator sequence Ni. The slew rate SR may be set to SR=20 (T/m/s), for example.

[0173] While the MR signal A obtained by the navigator sequence Ni is described referring to FIG. 30, the other navigator sequences N₂ to N_a are also expressed by the same sequence chart as that for the navigator sequence Ni. Therefore, again, the MR signal A is obtained when the other navigator sequences N₂ to N_a are performed. FIG. 31 shows MR signals A obtained by the navigator sequences Ni to N_a. In FIG. 31, subscripts " 1," "2," "3," "a-1," and "a" are added to symbol A to distinguish the MR signals A obtained by the navigator sequences Ni to N_a from one another.

[0174] The navigator sequences Ni to N_a are configured to have no gradient pulse applied during excitation, and moreover, to have a small value of the slew rate SR for the killer pulse K. Therefore, the navigator sequences Ni to N_a can sufficiently reduce noise during performance of the sequences as compared with those of a pencil-beam type.

[0175] In the second embodiment, a respiratory signal is acquired by performing the prescan PS shown in FIG. 31. Now the flow of imaging in the second embodiment will be described below.

[0176] FIG. 32 is a diagram showing the flow of imaging in the second embodiment. Since Steps ST1 and ST2 are the same as those in the first embodiment, their explanation will be omitted. After performing the localizer scan LS at Step ST2, the flow goes to Step ST3. [0177] At Step ST3, the operator operates the operating section 11, and inputs information required to define a range of the imaged body part in a main scan MS while referring to the image acquired in the localizer scan LS. Once the information has been input via the operating section 11, the range defining unit 921 (see FIG. 29) defines a range of the imaged body part based on the input information. FIG. 33 schematically shows the range of the imaged body part defined at Step ST3. After defining the range of the imaged body part, the flow goes to Step ST4.

[0178] At Step ST4, a prescan PS is performed. Now the prescan PS will be described below. The prescan PS will be explained referring to FIGS. 34 to 36.

[0179] First, as shown in FIG. 34, a navigator sequence Ni is performed. Since the navigator sequence Ni performs excitation using the non-selective RF pulse EXi (see FIG. 30), a wide range of a body part (torso including the liver and lungs, for example) may be excited by performing the navigator sequence Ni. An MR signal Ai generated from the excited body part is received by the receive coil apparatus 4 (see FIG. 1).

[0180] Since the receive coil apparatus 4 has the channels CHI to CH8, the MR signal Ai is received by each of the channels CHI to CH8. Signals received at the channels CHI to CH8 are transmitted to the receiver 8. The receiver 8 applies signal processing, such as

demodulation/detection, to the signals received from the channels. Therefore, by performing the navigator sequence Ni, navigator data containing information (respiratory information) on the MR signal Ai may be obtained on a channel-by-channel basis. The navigator data are schematically denoted here by symbols "An," "Ai₂," "Ai₃," "Ai₄," "Ai₅," "Ai₆," "An," and "Ai8." The navigator data An to Ais are supplied to the processing apparatus 9. [0181] The processing apparatus 9 determines a signal value of a respiratory signal based on the navigator data An to A₁₈. Now a scheme of determining a respiratory signal in the second embodiment will be described below.

[0182] Similarly to the first embodiment, in the second embodiment, the channel identifying unit 91 (see FIG. 29) refers to the database (see FIG. 19) before the prescan PS to identify the channels CHI, CH2, CH5, and CH6 lying near the edge El of the liver and the channels CH3, CH4, CH7, and CH8 disposed at positions away from the edge El. Then, the respiratory signal generating unit 93 (see FIG. 29) determines a signal value of the respiratory signal based on the navigator data An to A₁₈ obtained by performing the prescan PS. Specifically, the signal value of the respiratory signal is determined as follows.

[0183] The respiratory signal generating unit 93 first combines together only the navigator data An, An, A₁₅, and A₁₆ obtained by the channels CHI, CH2, CH5, and CH6. Here, the respiratory signal generating unit 93 combines the navigator data An, A , A₁₅, and A½ from the channels CHI, CH2, CH5, and CH6 together by adding them together. This gives combined data SYn. After acquiring the combined data SYn, the respiratory signal generating unit 93 time-integrates the combined data SYn to calculate an integral value Si.

[0184] The respiratory signal generating unit 93 also combines together only the navigator data A , A₁₄, An, and A₁₈ obtained by the channels CH3, CH4, CH7, and CH8. Here, the respiratory signal generating unit 93 combines the navigator data An, A₁₄, An, and A₁₈ obtained by the channels CH3, CH4, CH7, and CH8 together by adding them together. This gives combined data SYn. After acquiring the combined data SYn, the respiratory signal generating unit 93 time-integrates the combined data SYn to calculate an integral value S₂. [0185] Next, the respiratory signal generating unit 93 determines a ratio r between the integral values Si and S₂. In FIG. 34, this is represented as r=Si/S₂=ri. In the second embodiment, the ratio r between integral values is used as the signal value r of the respiratory signal for the subject.

[0186] After performing the navigator sequence Ni, a next navigator sequence N₂ is performed. FIG. 35 is an explanatory diagram for determining the signal value of a respiratory signal based on the navigator data obtained by the navigator sequence N₂.

[0187] In FIG. 35, the navigator data obtained by the navigator sequence N₂ are

schematically shown by symbols "A₂i," "A₂₂," "A₂₃," "A₂4," "A₂₅," "A₂₆," "A₂₇," and "A₂₈."

[0188] The respiratory signal generating unit 93 first combines only the navigator data A₂₁, A₂₂, A₂5, and A₂₆ obtained by the channels CHI, CH2, CH5, and CH6 together to obtain combined data SY₂₁. After acquiring the combined data SY21, the respiratory signal generating unit 93 time-integrates the combined data SY21 to calculate an integral value Si.

[0189] The respiratory signal generating unit 93 also combines only the navigator data A₂₃, A24, A27, and A₂₈ obtained by the channels CH3, CH4, CH7, and CH8 together to obtain combined data SY22. After acquiring the combined data SY22, the respiratory signal generating unit 93 time-integrates the combined data SY22 to calculate an integral value S₂.

[0190] Next, the respiratory signal generating unit 93 determines a ratio r between the integral values Si and S₂. In FIG. 35, the ratio calculated by performing the navigator sequence N2 is represented as r=S i/S2=T2. [0191] Similarly thereafter, the navigator sequences N₃ to N_a are performed. The respiratory signal generating unit 93 determines combined data for the navigator data from the channels CHI, CH2, CH5, and CH6, and combined data for the navigator data from the channels CH3, CH4, CH7, and CH8 to calculate respective integral values for the combined data, and calculates ratios r between the integral values. Therefore, by performing the navigator sequences Ni to N_a, a respiratory signal S_res8 may be obtained, as shown in FIG. 36.

[0192] After performing the prescan PS, the flow goes to Step ST5. At Step ST5, the window defining unit 94 (see FIG. 29) defines a window W based on the respiratory signal S_res8. The method of defining the window W is similar to that in the first embodiment. After defining the window W, the flow goes to Step ST6. At Step ST6, a main scan MS is performed.

[0193] FIG. 37 is a diagram explaining the main scan MS. In FIG. 37, sequences performed in the main scan MS, and a respiratory signal S_res9 obtained by performing the sequences are schematically shown.

[0194] In the main scan MS, navigator sequences Nb to N_c are first performed. When each of the navigator sequences Nb to N_c is performed, the respiratory signal generating unit 93 calculates a ratio between integral values to determine a signal value of a respiratory signal. In FIG. 37, the signal values of the respiratory signal determined by performing the navigator sequences Nb to N_c are denoted by symbols ¾," "rb₊i," "r_c."

[0195] When each of the navigator sequences Nb to N_c is performed, the deciding unit 95 (see FIG. 29) decides whether or not the signal value of the respiratory signal falls within the window W. Then, when the signal value has entered the inside of the window W from the outside of the window W, an imaging sequence DAQi for collecting imaging data is performed. In the present embodiment, the imaging sequence DAQi is a 3D excitation sequence for 3D- exciting the imaged body part (see FIG. 33) and collecting volume data from the imaged body part.

[0196] Referring to FIG. 37, the signal values n, to r_c-1 from the navigator sequences Nb to Nc-i fall outside of the window W. However, the signal value r_c from the navigator sequence N_c enters the inside of the window W. Therefore, the imaging sequence DAQi is performed immediately after the navigator sequence N_c.

[0197] After performing the imaging sequence DAQi, the navigator sequences N_c+i to Nd are performed. When each of the navigator sequences N is performed, the respiratory signal generating unit 93 determines a signal value of the respiratory signal. Then, when the signal value of the respiratory signal has entered the inside of the window W from the outside of the window W, an imaging sequence DAQ₂ for collecting imaging data is performed. Referring to FIG. 37, the signal values r_c+2 to ¾-i from the navigator sequences N_c+2 to Nd-i fall outside of the window W. However, the signal value rd from the navigator sequence Nd enters the inside of the window W. Therefore, the imaging sequence DAQ₂ is performed immediately after the navigator sequence Nd.

[0198] Similarly thereafter, the navigator sequence is repetitively performed until imaging data required in image reconstruction for the slices are acquired, and when the signal value of the respiratory signal has entered the window W, the imaging sequence is performed. In the second embodiment, it is considered that imaging data required in image reconstruction for the slices have been acquired by performing a sequence set DAQ_Z. Therefore, after performing the imaging sequence DAQ_Z, the main scan MS is terminated. [0199] In the second embodiment, combined data SYn, SY21, SY_ai are determined based on MR signals received at the channels CHI, CH2, CH5, and CH6, and combined data SY12, SY22, ..·, SY_a2 are determined based on MR signals received at the channels CH3, CH4, CH7, and CH8. Then, integral values for the combined data are calculate to determine ratios r between the integral values as signal values of a respiratory signal. By thus calculating the ratio r between the integral values, the impact of steep attenuation of an MR signal encountered in a period of time from immediately after the start of the prescan PS and until the MR signal reaches a steady state may be reduced. Therefore, a high-quality respiratory signal may be obtained from immediately after the start of the prescan PS.

[0200] In the second embodiment, after determining the respiratory signal S_res8 by the prescan PS, a window W is defined, and in the main scan MS, an imaging sequence is performed when the signal value of the respiratory signal has entered the window W. Therefore, variability of the phase of respiration when performing the imaging sequences DAQi to DAQ_Z may be fully decreased in the main scan MS, so that an image having reduced body motion artifacts may be obtained.

[0201] Moreover, in the prescan PS in the second embodiment, navigator sequences Ni to N_a are performed with no gradient pulse applied in excitation. Therefore, noise during performance of the navigator sequences may be fully reduced.

[0202] In the second embodiment, the navigator sequence has a killer pulse K (see FIG. 30). Therefore, transverse magnetization, which causes an error in signal value of the respiratory signal, may be canceled before starting a next navigator sequence, so that it is possible to obtain a higher-quality respiratory signal. [0203] It should be noted that in the second embodiment, to reduce noise during performance of the navigator sequences, the slew rates SR for the rise time and fall time of the killer pulse K is decreased. However, a killer pulse having a sinusoidal ramp (Reference: F. Hennel, F. Girard, and T. Loenneker, "'Silent' MRI With Soft Gradient Pulses," Magnetic Resonance in Medicine, 42:6-10 (1999)) may be employed in place of that having a small slew rate SR.

[0204] In a third embodiment, a method of determining a respiratory signal using sequences according to the DC self-navigated method, as in the first embodiment. However, while the first embodiment addresses a case in which the respiratory signal is determined focusing upon the positional relationship between the edge El of the liver and channels in a coil, the third embodiment addresses a case in which the respiratory signal is determined focusing upon the positional relationship between the edge El of the liver and slices. The hardware configuration of the MR apparatus in the third embodiment is the same as that in the first embodiment.

[0205] FIG. 38 is a diagram explaining processing the processing apparatus 9 executes in the third embodiment. As compared with the first embodiment, the third embodiment comprises slice identifying unit 911 in place of the channel identifying unit 91, and the other units are the same as those in the first embodiment. Therefore, in explaining the processing apparatus 9 in the third embodiment, the slice identifying unit 911 will be described and description of the other units will be omitted.

[0206] The slice identifying unit 91 identifies, from among a plurality of defined slices, u slices defined near the edge El of the liver (see FIGS. 3 and 4), and v slices defined at positions farther away from the edge El of the liver than the u slices are. [0207] The processing apparatus 9 is an example for configuring the slice identifying unit 911 to deciding unit 95, and it functions as these units by loading programs stored in the storage section 10.

[0208] FIG. 39 is a diagram schematically showing an imaged body part, and slices defined in the imaged body part in the third embodiment. In the third embodiment, the imaged body part is the liver, and the slices are defined to transversely cross the liver. The third embodiment shows a case in which axial slices Xi to X20 are defined as the slices.

[0209] Next, a basic concept of the method of generating a respiratory signal in the third embodiment will be described referring to FIG. 40. In FIG. 40, the positional relationship between the axial slices Xi to X20 and the liver is schematically shown. The liver when the subject expires is indicated by a solid line, while the liver when the subject inspires is indicated by a dashed line. Moreover, axial slices X3 and X13 are also shown in a plan view in FIG. 40.

[0210] First, let us consider the axial slice X3 lying near the edge El of the liver. When the subject expires, the edge El of the liver moves in the S-direction, so that the liver tends to have an increased cross-sectional area in the axial slice X3. Since the liver is a high-signal source as compared with the lungs, the signal intensity of MR signals the channels receive from the axial slice X3 increases as the cross-sectional area of the liver that is a high-signal source increases in the axial slice X3. Therefore, when the subject expires, the signal intensity of MR signals received by the channels tends to increase. When the subject inspires, on the other hand, the edge El of the liver moves in the I-direction, so that the liver tends to have a decreased cross- sectional area in the axial slice X3. Therefore, when the subject inspires, the signal intensity of MR signals the channels receive from the axial slice X3 tends to decrease. Thus, the signal intensity of MR signals the channels receive from the axial slice X₃ increases when the subject expires, while the signal intensity of MR signals the channels receive from the axial slice X₃ decreases when the subject inspires.

[0211] Next, consider the axial slice X₁₃ lying at a position farther away from the edge El of the liver than the axial slice X₃ does. The axial slice X₁₃ exhibits a tendency opposite to that of the axial slice X₃. Specifically, the signal intensity of MR signals the channels receive from the axial slice X₁₃ decreases when the subject expires, while the signal intensity of MR signals the channels receive from the axial slice X₁₃ increases when the subject inspires.

[0212] In the preceding description, the slices X₃ and X₁₃ are taken to explain the difference in time of increase and decrease of the signal intensity of MR signals received by the channels for convenience of explanation. However, the slices Xi to X20 may be separately considered as a group Gu and a group G_v, the group G_u consisting of u slices (u=10, that is, slices Xi to X10 here) defined adjacent to the edge El of the liver, and the group G_v consisting of v slices (v=10, that is, slices X11 to X20 here) defined at positions away from the edge El of the liver, wherein a major portion of the slices Xi to X10 defined adjacent to the edge El of the liver may be considered similarly to the slice X₃, while a major portion of the slices Xn to X20 defined at positions away from the edge El of the liver may be considered similarly to the slice X₁₃.

[0213] From the considerations above, it can be seen that the times of the increase and decrease of MR signals the channels receive from (a major portion of) the axial slices Xi to X10 are the reverse of those of MR signals the channels receive from (a major portion of) the axial slices Xn to X20. Therefore, when all MR signals obtained from all the slices Xi to X20 are added together, the amplitude of the respiratory signal cannot be increased, resulting in a problem that it is difficult to obtain a respiratory signal fully reflecting subject's respiratory motion.

[0214] Then, as an example, a method of generating a respiratory signal may be

contemplated to involve selecting slices in which received signals at the channels

increase/decrease at the same timing from among the slices Xi to X20, and generating a respiratory signal based on the signals received in the selected slices. As described above, MR signals obtained from the u slices Xi to X10 of the slices Xi to X20 that are defined to lie near the edge El of the liver tend to increase/decrease approximately at the same timing. Accordingly, it may be contemplated to select the slices Xi to X10 lying adjacent to the edge El of the liver from among the slices Xi to X20, and generate a respiratory signal based on signals received in the selected slices Xi to X10. The method, however, poses a problem that the signal value of the respiratory signal is not stabilized in a certain period of time from the start of the prescan PS due to attenuation of the MR signal which is encountered until the MR signal reaches a steady state (see FIG. 51). Accordingly, the inventor of the present application has made an intensive study and found that a respiratory signal may be generated such that the impact of attenuation of the MR signal until it reaches a steady state is reduced by using MR signals obtained from the slices X11 to X20, in addition to those obtained from the slices Xi to X10. Now a scheme of determining a respiratory signal in the third embodiment will be described referring to the flow in FIG. 9.

[0215] Since Steps ST1 and ST2 are the same as those in the first embodiment, their explanation will be omitted. After performing the localizer scan at Step ST2, the flow goes to Step ST3. At Step ST3, slices are defined. In the third embodiment, axial slices Xi to X20 are defined as the slices, as shown in FIG. 39. [0216] After defining the slices, the flow goes to Step ST4. At Step ST4, a prescan PS is performed.

[0217] FIG. 41 is a diagram explaining the prescan PS. In FIG. 41, the prescan PS is illustrated separately in a plurality of periods of time Pi to P_a. In each period of time is performed a sequence set including sequences AXi to AX20 for collecting MR signals from the axial slices Xi to X20 (see FIG. 39) according to a multi-slice method. Each period of time represents the repetition time TR. FIG. 41 shows the plurality of sequences AXi to AX20 performed in a period of time Pi of the periods of time Pi to P_a. The sequences AXi to AX20 performed in the prescan PS are those according to the DC self-navigated method as in the first embodiment (see FIG 7; however, note that Gx and Gz change places). Therefore, by performing the sequences AXi to AX20, navigator signals (ai to a2o) and imaging signals (bi to b2o) are collected from the axial slices Xi to X20.

[0218] After performing the sequences AXi to AX20 in the period of time Pi, the sequences AXi to AX20 are also performed in a next period of time P2. Similarly thereafter, the sequences AXi to AX20 are repetitively performed. Therefore, the sequences AXi to AX20 are performed in each of the periods of time Pi to P_a.

[0219] In the third embodiment, a respiratory signal is acquired by performing the prescan PS. Now a method of acquiring a respiratory signal by performing the prescan PS will be described below.

[0220] FIG. 42 is an explanatory diagram for performing the sequences AXi to AX20 in the prescan PS. First, in the period of time Pi, the sequence AXi is performed. By performing the sequence AXi, a navigator signal ai and an imaging signal bi are collected. The navigator signal ai and imaging signal bi are received at the receive coil apparatus 4.

[0221] Since the receive coil apparatus 4 has the channels CHI to CH8, the navigator signal ai and imaging signal bi are received at each of the channels CHI to CH8, and transmitted to the receiver 8. The receiver 8 applies signal processing, such as demodulation/detection, to the signals received from each of the channels CHI to CH8. Thus, navigator data An to A₁₈ containing information (respiratory information) on the navigator signal ai and imaging data Bn to Bis containing information (image information) on the imaging signal bi are obtained.

[0222] Similarly thereafter, sequences AX₂ to AX₂o are performed for collecting navigator signals and imaging signals from the axial slices X₂ to X₂o. Therefore, navigator data An to A₂os and imaging data Bn to B₂os are obtained by performing the prescan PS.

[0223] After performing the sequences AX₂ to AX₂o in the period of time Pi, a signal value of a respiratory signal in the period of time Pi is determined. Now a method of determining the signal value of a respiratory signal in the period of time Pi will be described below.

[0224] FIG. 43 is a diagram explaining the method of determining the signal value of a respiratory signal. In the third embodiment, before performing the prescan PS, the slice identifying unit 911 (see FIG. 38) identifies u (=10) slices Xi to Xio (see FIG. 40) defined adjacent to the edge El of the liver, and v (=10) slices Xn to X₂o (see FIG. 40) defined at positions away from the edge El of the liver.

[0225] Then, the respiratory signal generating unit 93 combines together the navigator data An to Aio8 obtained on a channel-by-channel basis from the slices Xi to Xio. Here, the respiratory signal generating unit 93 combines these navigator data An to Aios together by adding them together. This gives combined data SYi.

[0226] After acquiring the combined data SYi, the respiratory signal generating unit 93 determines a feature quantity of the navigator signals that the channels CHI to CH8 have collected from the slices Xi to Xio based on the combined data SYi. In the third embodiment, the combined data SYi is time-integrated to thereby calculate an integral value Si, which is determined as a feature quantity of the navigator signals that the channels CHI to CH8 have collected from the slices Xi to Xio.

[0227] Next, the respiratory signal generating unit 93 combines the navigator data Am to A208 obtained on a channel-by-channel basis from the v (=10) slices Xn to X20 defined at positions away from the edge El of the liver. Here, the respiratory signal generating unit 93 combines these navigator data Am to A208 together by adding them together. This gives combined data SY₂.

[0228] After acquiring the combined data SY₂, the respiratory signal generating unit 93 determines a feature quantity of the navigator signals that the channels CHI to CH8 have collected from the slices Xn to X20 based on the combined data SY₂. In the third embodiment, the combined data SY₂ is time-integrated to thereby calculate an integral value S₂, which is determined as a feature quantity of the navigator signals that the channels CHI to CH8 have collected from the slices Xn to X20.

[0229] Next, the respiratory signal generating unit 93 determines a ratio r between the integral values Si and S₂. In FIG. 43, this is represented as r=Si/S₂=ri. In the third embodiment, the ratio r between integral values is adopted as the signal value r of the respiratory signal for the subject.

[0230] Similarly thereafter, the sequences AXi to AX20 are performed in each of the periods of time P₂ to P_a to calculate the ratio r between integral values of combined data. Thus, signal values r=n, r₂, r₃, r_a of the respiratory signal in the periods of time Pi, P₂, P₃, P_a may be determined (see FIG. 44).

[0231] FIG. 44 is a diagram schematically showing a respiratory signal Sresii determined according to the method in the third embodiment. The respiratory signal generating unit 93 determines an integral value S i for the combined data SYi and an integral value S₂ for the combined data SY₂, and determines a ratio r between the integral values as a signal value of the respiratory signal. By thus calculating a ratio r between integral values, the impact of steep attenuation of an MR signal encountered in a period of time from immediately after the start of the prescan PS and until the MR signal reaches a steady state may be reduced. Therefore, a stable respiratory signal may be generated from immediately after the start of the prescan PS. Moreover, by calculating the ratio r between integral values, the amplitude of the respiratory signal may be increased.

[0232] While the sequences used in the prescan PS have gradient pulses Gyl and Gy2 in the phase encoding direction (see FIG. 7), the prescan PS is not a scan performed for determining images in the slices Xi to X20. Therefore, the values of the magnetic field intensity G of the gradient pulses Gyl and Gy2 in the phase encoding direction may be unchanged during the prescan PS. Typically the prescan PS may be performed with the magnetic field intensity G set to G=0. [0233] After determining the respiratory signal Sresi i, the flow goes to Step ST5. At Step ST5, based on the respiratory signal S_res4, a window is defined for deciding whether to accept imaging data acquired in a main scan MS, which will be discussed later (Step ST6), as data for image reconstruction or to discard them. Now a method of defining the window will be described below.

[0234] First, the window defining unit 94 (see FIG. 38) determines a signal value

corresponding to a phase of respiration representing end expiration based on the signal value of the respiratory signal resl l -

[0235] FIG. 45 is an explanatory diagram for determining a signal value corresponding to the phase of respiration representing end expiration. In FIG. 45, a positional relationship between the slices Xi to X20 and the liver is schematically shown. The liver when the subject expires is indicated by a solid line, while the liver when the subject inspires is indicated by a dashed line.

[0236] Let us first consider a case in which the subject expires. When the subject expires, the edge El of the liver moves in the S-direction, so that the cross-sectional area of the liver increases in (a major portion of) the slices Xi to Xio adjacent to the edge El of the liver.

Therefore, the signal intensity of MR signals collected from (a major portion of) the slices Xi to Xio increases, resulting in a large integral value Si for the combined data. On the other hand, the cross- sectional area of the liver tends to decrease in (a major portion of) the slices Xn to X20 at positions away from the edge El of the liver. Therefore, the signal intensity of MR signals collected from (a major portion of) the slices Xn to X20 decreases, resulting in a small integral value S2 for the combined data. Therefore, when the subject expires, the ratio r between the integral values has a large value. [0237] On the other hand, when the subject inspires, the ratio r between the integral values has a small value, as opposed to the case in which the subject expires. Accordingly, in the third embodiment, a local maximum (which is a signal value when the signal value of a respiratory signal changes from an increase to a decrease) of the respiratory signal is decided to be a signal value corresponding to a phase of respiration representing end expiration. FIG. 46 schematically shows a signal value rx corresponding to the phase of respiration representing end expiration determined by the window defining unit 94.

[0238] After determining the signal value rx corresponding to the phase of respiration representing end expiration, the window defining unit 94 defines a window W based on the signal value rx. The same method of defining a window W as that in the first embodiment may be used. After defining the window W, the flow goes to Step ST6. At Step ST6, a main scan MS for acquiring images in the axial slices Xi to X20 (see FIG. 39) is performed.

[0239] FIG. 47 is a diagram explaining the main scan MS. The main scan MS performs sequences AXi to AX20 in periods of time Pi to Pb. The sequences AXi to AX20 performed in the main scan MS are sequences according to the DC self-navigated method, as with those performed in the prescan PS. Therefore, in the main scan MS, again, navigator signals (ai to a2o) and imaging signals (bi to b2o) are collected by performing the sequences AXi to AX20. Now the main scan MS will be particularly described below.

[0240] FIG. 48 is an explanatory diagram for performing the sequences AXi to AX20 in the main scan MS. First, a sequence AXi is performed in the period of time Pi. By performing the sequence AXi, a navigator signal ai and an imaging signal bi are collected. The magnetic field intensity G of the gradient pulses in the phase encoding direction for the sequence AXi is set to a value G=0 for obtaining imaging data on a line at ky=0. The navigator signal ai and imaging signal bi are received at the receive coil apparatus 4.

[0241] The navigator signal ai is received at each of the channels CHI to CH8 in the receive coil apparatus 4, and transmitted to the receiver 8. The receiver 8 applies signal processing, such as demodulation/detection, to the signals received from the receive coil apparatus 4, and outputs navigator data An to A₁₈ containing information (respiratory information) on the navigator signal ai and imaging data Bn to Bis containing information (image information) on the imaging signal bi to the processing apparatus 9.

[0242] After performing the sequence AXi, and similarly thereafter, sequences AX₂ to AX₂o for acquiring images in the axial slices X₂ to X₂o are sequentially performed. The magnetic field intensity G of the gradient pulses in the phase encoding direction for the sequences AX₂ to AX₂o is set to a value G=0 for obtaining imaging data on a line at ky=0. Therefore, imaging data on a line at ky=0 is acquired from each of the axial slices Xi to X₂o in the period of time Pi.

[0243] After performing the sequences AXi to AX₂o in the period of time Pi, signal values of a respiratory signal in the period of time Pi are determined as follows.

[0244] The respiratory signal generating unit 93 combines together the navigator data An to Aio8 obtained on a channel-by-channel basis from the slices Xi to Xio. Here, the respiratory signal generating unit 93 combines these navigator data An to Aios together by adding them together. This gives combined data SYi.

[0245] After acquiring the combined data SYi, the respiratory signal generating unit 93 determines a feature quantity of the navigator signals that the channels CHI to CH8 have collected from the slices Xi to Xio based on the combined data SYi. In the third embodiment, the combined data SYi is time-integrated to thereby calculate an integral value Si, which is determined as a feature quantity of the navigator signals that the channels CHI to CH8 have collected from the slices Xi to Xio.

[0246] Next, the respiratory signal generating unit 93 combines together the navigator data Am to A208 obtained on a channel-by-channel basis from the slices Xn to X20. Here, the respiratory signal generating unit 93 combines these navigator data Am to A208 together by adding them together. This gives combined data SY₂.

[0247] After acquiring the combined data SY₂, the respiratory signal generating unit 93 determines a feature quantity of the navigator signals that the channels CHI to CH8 have collected from the slices Xn to X20 based on the combined data SY₂. In the third embodiment, the combined data SY₂ is time-integrated to thereby calculate an integral value S₂, which is determined as a feature quantity of the navigator signal that the channels CHI to CH8 have collected from the slices Xn to X20.

[0248] After calculating these integral values Si and S₂, the respiratory signal generating unit 93 calculates a ratio r between the integral values Si and S₂. In FIG. 48, this is represented as r=Si/S₂=ri. The ratio r=n between the integral values is adopted as the signal value r of the respiratory signal for the subject in the period of time Pi.

[0249] After determining the ratio r between the integral values, the deciding unit 95 decides whether or not to use the imaging data Bn to B₂os collected in the period of time Pi as data for image reconstruction based on the ratio r=n between integral values for the respiratory signal. Referring to FIG. 48, the signal value (ratio between the integral values) r=n in the period of time Pi does not fall within the window W, and therefore, the imaging data Bn to B208 collected in the period of time Pi are discarded. After performing the sequences in the period of time Pi, the process moves to a period of time P₂.

[0250] In the period of time P₂, data discarded in the period of time Pi are re-acquired. FIG. 49 is an explanatory diagram for re-acquiring data in the period of time P₂.

[0251] In the period of time P₂, the magnetic field intensity G of the gradient pulses in the phase encoding direction for the sequence is set to a value G=0 for obtaining imaging data on a line at ky=0, as in the period of time Pi. Therefore, in the period of time P₂, imaging data on a line at ky=0 is obtained from each of the axial slices Xi to X20.

[0252] After performing the sequences AXi to AX20 in the period of time P₂, the respiratory signal generating unit 93 generates combined data SYi and SY₂, and calculates an integral value Si for the combined data SYi and an integral value S₂ for the combined data SY₂. The respiratory signal generating unit 93 then calculates a ratio r between the integral values. In FIG. 49, the ratio r in the period of time P₂ is represented as r=Si/S₂=r₂. Therefore, r=r₂ is used as the signal value of the respiratory signal in the period of time P₂.

[0253] Next, the deciding unit 95 decides whether or not the signal value of the respiratory signal falls within the window W. Referring to FIG. 49, the signal value (ratio between the integral values) r=r 2 in the period of time P₂ falls within the window W. Therefore, the deciding unit 95 decides that the imaging data Bn to B₂os collected in the period of time P₂ as data for image reconstruction. [0254] Similarly thereafter, the sequences AXi to AX20 are performed in each period of time, and whether or not the signal value of the respiratory signal falls within the window W in the period of time is decided. In the case that the signal value does not fall within the window W, the sequences AXi to AX20 are performed in a next period of time without changing the value of the magnetic field intensity G of the gradient pulses in the phase encoding direction. On the other hand, in the case that the signal value falls within the window W, the value of the magnetic field intensity G for the gradient pulses in the phase encoding direction is changed, and the sequences AXi to AX20 for acquiring imaging data on another line in ky is performed in the next period of time. The sequence set of the sequences AXi to AX20 is repetitively performed until all data in k-space required in image reconstruction are acquired. Once all data in k-space required in image reconstruction have been acquired, images in the axial slices Xi to X20 (see FIG. 39) are reconstructed, and the flow of the third embodiment is terminated.

[0255] In the third embodiment, combined data SYi is determined based on MR signals generated from the axial slices Xi to X10, and combined data SY2 is determined based on MR signals generated from the axial slices Xn to X20. Then, an integral value Si for the combined data SYi and an integral value S2 for the combined data SY2 are determined, and a ratio r between the integral values is determined as a signal value of a respiratory signal. By thus calculating the ratio r between integral values, the impact of steep attenuation of an MR signal encountered in a period of time from immediately after the start of the prescan PS and until the MR signal reaches a steady state may be reduced. Therefore, a stable respiratory signal may be generated from immediately after the start of the prescan PS. Moreover, by calculating the ratio r between integral values, the amplitude of the respiratory signal may be increased. [0256] In the third embodiment, as the u slices of the twenty slices Xi to X20 that are defined near the edge El of the liver, u=10, that is, ten slices Xi to X10 are selected (see FIG. 40).

However, u may be u<10 or u>10 insofar as a high-quality respiratory signal can be generated. For example, as the u slices defined near the edge El of the liver, u=5, that is, five slices (for example, slices Xi to X5) may be selected, or u=l 1, that is, eleven slices (for example, slices Xi to Xn) may be selected. Moreover, as the u slices defined near the edge El of the liver, u=l, that is, a single slice (for example, slice X3) may be selected.

[0257] Moreover, in the third embodiment, as the v slices of the twenty slices Xi to X20 that are defined at positions away from the edge El of the liver, v=10, that is, ten slices Xn to X20 are selected (see FIG. 40). However, v may be v<10 or v>10 insofar as a high-quality respiratory signal can be generated. For example, as the v slices defined at positions away from the edge El of the liver, v=5, that is, five slices (for example, slices Xn to X15) may be selected, or v=l 1, that is, eleven slices (for example, slices X10 to X20) may be selected. Moreover, as the v slices defined at positions away from the edge El of the liver, v=l, that is, a single slice (for example, slice X13) may be selected.

[0258] In the third embodiment, the integral value for combined data is calculated as a feature quantity for MR signals collected from slices. However, another feature quantity may be determined in place of the integral value insofar as a respiratory signal can be generated. For example, the maximum of combined data may be determined as the feature quantity. In the case that the maximum of combined data is determined, the ratio between the maxima for the combined data is used as a signal value of a respiratory signal. [0259] In the first to third embodiments, the combined data is obtained by adding navigator data together. However, the combination of navigator data is not limited to addition, and the combined data may be obtained by, for example, weighted addition of navigator data, or by multiplying navigator data.

[0260] In the first to third embodiments, a respiratory signal is exemplified as a body motion signal for a subject. However, the present invention is not limited to acquisition of the respiratory signal. For example, in the case that a heart is imaged, it is possible to acquire a heartbeat signal containing information on cardiac pulsatility by identifying, from among a plurality of slices defined in the heart, those defined near an edge of the heart and those defined at positions away from the edge of the heart. It is also possible to acquire a heartbeat signal containing information on cardiac pulsatility by identifying, from among the channels CHI to CH8, those disposed near the edge of the heart and those disposed at positions away from the edge of the heart.

[0261] In the first and second embodiments, information in the database (see FIG. 19) is referred to thereby identify n channels (CHI, CH2, CH5, and CH6) disposed near the edge El of the liver and m channels (CH3, CH4, CH7, and CH8) disposed at positions away from the edge El of the liver than the n channels are. However, rather than registering channels in the database, a channel identifying scan for identifying, from among the channels CHI to CH8, channels CHI, CH2, CH5, and CH6 disposed near the edge El of the liver and channels CH3, CH4, CH7, and CH8 disposed at positions away from the edge El of the liver may be performed, and based on data obtained in the scan, the channels CHI, CH2, CH5, and CH6 disposed near the edge El of the liver and the channels CH3, CH4, CH7, and CH8 disposed at positions away from the edge El of the liver may be distinguished from one another. FIG. 50 is a diagram showing an example for performing the channel identifying scan according to a fourth embodiment. FIG. 50 shows a case in which a channel identifying scan ES is performed between the localizer scan LS and prescan PS. As described regarding the first embodiment, the channels CHI, CH2, CH5, and CH6 and the channels CH3, CH4, CH7, and CH8 are in a relationship such that increase and decrease of the signal intensity of MR signals that they receive are the reverse of each other, so that the channels (CHI, CH2, CH5, and CH6) disposed near the edge El of the liver, and the channels (CH3, CH4, CH7, and CH8) disposed at positions away from the edge El of the liver may be distinguished from one another based on the signal intensity at each channel. For example, a profile representing a change of the signal intensity in the z-direction (Si-direction) may be generated on a channel-by-channel basis, and by analyzing a channel-to-channel difference in the profile, the channels (CHI, CH2, CH5, and CH6) and channels (CH3, CH4, CH7, and CH8) may be distinguished from one another.

[0262] As such, a channel identifying scan may be performed rather than registering channels in a database. In the case that the channel identifying scan ES is performed, even when the array coils 4a and 4b are somewhat offset from their ideal positions toward the S- or I-side with respect to the edge El of the liver, channels lying on the S-side and those lying on the I-side may be distinguished from among the channels CHI to CH8. Therefore, even when the array coils 4a and 4b are somewhat offset from their ideal positions toward the S- or I-side with respect to the edge El of the liver, a high-quality respiratory signal may be generated.

[0263] Moreover, in the case that the channel identifying scan ES is performed, it is not necessary to register channels to be disposed near the edge El of the liver in a database for each coil to be used in imaging, so that the workload in maintenance of the database may be reduced.

Claims

CLAIMS What is claimed is:

1. A magnetic resonance apparatus comprising:

a scanning section for performing a first scan for generating a first MR signal from a first body part including a moving body part of a subject, said first MR signal containing information on body motion of said subject;

a coil apparatus having a plurality of channels for receiving said first MR signal generated by said first scan;

a channel identifying unit for identifying n (n > 1) channels and m (m > 1) channels of said plurality of channels, said n channels being disposed near an edge of said moving body part, and said m channels being disposed at positions farther away from said edge of said moving body part than said n channels are; and

a unit for determining a signal value of a body motion signal representing body motion of said subject, said unit determining a first feature quantity of said first MR signal received by said n channels and a second feature quantity of said first MR signal received by said m channels, and determining said signal value of said body motion signal based on said first feature quantity and said second feature quantity.

2. The magnetic resonance apparatus as recited in claim 1, further comprising: a data generating unit for generating navigator data containing said information on body motion of said subject on a basis of said channel-by-channel based on said first MR signal received by each of said plurality of channels; wherein said unit for determining a signal value of a body motion signal determines said first feature quantity based on said navigator data obtained by said n channels and determines said second feature quantity based on said navigator data obtained by said m channels.

3. The magnetic resonance apparatus as recited in claim 2, wherein n > 2 and m > 2; and said unit for determining a signal value of a body motion signal:

combines said navigator data obtained by said n channels to thereby generate first combined data, and determines said first feature quantity based on said first combined data; and combines said navigator data obtained by said m channels to thereby generate second combined data, and determines said second feature quantity based on said second combined data.

4. The magnetic resonance apparatus as recited in claim 3, wherein said unit for determining a signal value of a body motion signal:

determines an integral value obtained by time-integrating said first combined data as said first feature quantity; and

determines an integral value obtained by time-integrating said second combined data as said second feature quantity.

5. The magnetic resonance apparatus as recited in claim 2, wherein a plurality of slices are defined in said first body part; and said scanning section performs in said first scan a sequence set including a plurality of sequences for generating said first MR signal from said plurality of slices.

6. The magnetic resonance apparatus as recited in claim 5, wherein each of said plurality of sequences is a sequence for generating, in addition to said first MR signal, a second MR signal containing image information for the subject from said slice; and said data generating unit determines imaging data containing said image information for the subject on a basis of said channel-by-channel based on said second MR signal.

7. The magnetic resonance apparatus as recited in claim 2, wherein said first scan is a scan for obtaining said body motion signal; and said scanning section performs a second scan for acquiring an image in each of said plurality of slices.

8. The magnetic resonance apparatus as recited in claim 1, wherein said channel identifying unit identifies said n channels and said m channels based on a database containing information for identifying said n channels and said m channels from among said plurality of channels.

9. The magnetic resonance apparatus as recited in claim 1, wherein said scanning section performs a third scan for identifying said n channels and said m channels before said first scan.

10. The magnetic resonance apparatus as recited in claim 1, wherein said first MR signal contains information on data at a center of k-space.

11. A magnetic resonance apparatus comprising:

a scanning section for performing a first scan for generating a first MR signal from each of a plurality of slices defined in a first body part including a moving body part of a subject, said first MR signal containing information on body motion of said subject;

a slice identifying unit for identifying u (u > 1) slices and v (v > 1) slices of said plurality of slices, said u slices being defined near an edge of said moving body part, and said v slices being defined at positions farther away from said edge of said moving body part than said u slices are; and

a unit for determining a signal value of a body motion signal representing body motion of said subject, said unit determining a first feature quantity of said first MR signal obtained from said u slices and a second feature quantity of said first MR signal obtained from said v slices, and determining said signal value of said body motion signal based on said first feature quantity and said second feature quantity.

12. The magnetic resonance apparatus as recited in claim 11, further comprising: a data generating unit for generating navigator data containing said information on body motion of said subject on a basis of said slice-by- slice based on said first MR signal generated from each of said plurality of slices;

wherein said unit for determining a signal value of a body motion signal determines said first feature quantity based on said navigator data in said u slices and determines said second feature quantity based on said navigator data in said v slices.

13. The magnetic resonance apparatus as recited in claim 12, wherein u > 2 and v > 2; and said unit for determining a signal value of a body motion signal:

combines said navigator data in said u slices to thereby generate first combined data, and determines said first feature quantity based on said first combined data; and

combines said navigator data in said v slices to thereby generate second combined data, and determines said second feature quantity based on said second combined data.

14. The magnetic resonance apparatus as recited in claim 13, wherein said unit for determining a signal value of a body motion signal: determines an integral value obtained by time-integrating said first combined data as said first feature quantity; and

determines an integral value obtained by time-integrating said second combined data as said second feature quantity.

15. The magnetic resonance apparatus as recited in claim 12, wherein said scanning section performs in said first scan a sequence set including a plurality of sequences for generating said first MR signal from said plurality of slices.

16. The magnetic resonance apparatus as recited in claim 15, wherein each of said plurality of sequences is a sequence for generating, in addition to said first MR signal, a second MR signal containing image information for the subject from said slice; and said data generating unit determines imaging data containing said image information for the subject on a basis of said channel-by-channel based on said second MR signal.

17. The magnetic resonance apparatus as recited in claim 12, wherein said first scan is a scan for obtaining said body motion signal; and said scanning section performs a second scan for acquiring an image in each of said plurality of slices.

18. The magnetic resonance apparatus as recited in claim 17, wherein said unit for determining a signal value of a body motion signal determines a signal value of said body motion signal based on a ratio between said first feature quantity and said second feature quantity.

19. The magnetic resonance apparatus as recited in claim 18, wherein said body motion signal is a respiratory signal or a heartbeat signal.

20. The magnetic resonance apparatus as recited in claim 19, wherein said moving body part includes a liver; and said edge of said moving body part is an edge of the liver adjacent to lungs.

21. A program applied to a magnetic resonance apparatus comprising scanning section for performing a first scan for generating a first MR signal from a first body part including a moving body part of a subject, said first MR signal containing information on body motion of said subject, and a coil apparatus having a plurality of channels for receiving said first MR signal generated by said first scan, said program being for causing a computer to execute: a channel identifying process of identifying n (n>=l) channels and m (m>=l) channels of said plurality of channels, said n channels being disposed near an edge of said moving body part, and said m channels being disposed at positions farther away from said edge of said moving body part than said n channels are; and

a process of determining a first feature quantity of said first MR signal received by said n channels and a second feature quantity of said first MR signal received by said m channels, and determining a signal value of a body motion signal representing body motion of said subject based on said first feature quantity and said second feature quantity.

22. A program applied to a magnetic resonance apparatus for performing a first scan for generating a first MR signal from each of a plurality of slices defined in a first body part including a moving body part of a subject, said first MR signal containing information on body motion of said subject, said program being for causing a computer to execute:

a slice identifying process of identifying u (u > 1) slices and v (v > 1) slices of said plurality of slices, said u slices being defined near an edge of said moving body part, and said v slices being defined at positions farther away from said edge of said moving body part than said u slices are; and

a process of determining a first feature quantity of said first MR signal obtained from said u slices and a second feature quantity of said first MR signal obtained from said v slices, and determining a signal value of a body motion signal representing body motion of said subject based on said first feature quantity and said second feature quantity.