EP2049017A2

EP2049017A2 - Device and method for determining the optimum image reconstruction phase for objects moving quasy periodically

Info

Publication number: EP2049017A2
Application number: EP07801594A
Authority: EP
Inventors: Dirk Ertel; Marc Kachelriess
Original assignee: VAMP Verfahren und Apparate der Medizinischen Physik GmbH
Current assignee: CT Imaging GmbH
Priority date: 2006-08-10
Filing date: 2007-08-10
Publication date: 2009-04-22
Also published as: WO2008017493A3; DE102006037601A1; WO2008017493A2

Abstract

The invention relates to a device and a method for determining the optimum image reconstruction phase for objects moving quasi periodically, particularly with regard to the computer tomography (CT) of the heart. One task of the present invention is to enable an automated determining of the optimum image reconstruction phase with regard to the reduction of movement factors. This task is solved by a method for illustration of an object moving quasi periodically using an image recording system (1), particularly a computer tomograph, using a phase-correlated reconstruction technique, in which an optimum image reconstruction phase is determined using a movement function (8) of the object.

Description

description

Apparatus and method for determining the optimal image reconstruction phase for quasi-periodically moving objects

The invention relates to a device and a method for determining the optimal image reconstruction phase for quasi-periodically moving objects, in particular with regard to X-ray computed tomography (CT) of the heart.

The possibility of representing moving objects motionless in a quasi-momentary state is highly dependent on the temporal resolution of the recording system. In the context of this description, the recording by computer tomographs (CT, micro-CT, C-arm CT) is explained, wherein the proposed technique is applicable to any other recording system with multi-dimensional data. In particular, in high-resolution clinical CT, there is the difficulty that movement of organs, e.g. Heart, lungs, image blurring revealed.

Especially in clinical CT, the representation of the heart can lead to characteristic movement artifacts in the reconstructed images. Under the assumption of a quasi-periodic object movement, however, the reconstruction algorithm can be synchronized with the object movement. In this case, only projection data from the same movement phases (preferably those with minimal movements of the object to be examined) for a

Image reconstruction is used and the remaining data is not taken into account or with reduced weight. This can increase the temporal resolution of the imaging system and the objects are displayed in a quasi-static state.

In human cardiac CT, the use of a simultaneous recorded electrocardiogram (ECG) as

Synchronization signal most widely used for image reconstruction. Moreover, it is known to use the kymogram function, which is e.g. represents the time-dependent centroid movement of a slice under study to use as a synchronization signal.

Irrespective of the synchronization signal used, periodically recurring object states are detected for the image reconstruction. Only the projection data within a temporal window around the defined object state are used for a phase-correlated image reconstruction and thus provide a picture contribution, the remaining projection data are disregarded. This approach is common to all common reconstruction techniques, such as analytical filter protection (PBP) or statistical algebraic reconstruction technique (ART).

Because some of the projection data go unnoticed in image reconstruction, tube current modulation (TCM) can reduce the dose burden on a cardiac CT scan, with CT's tube current in the high-motion cardiac phases providing low image contributions However, this requires an availability of the synchronization signal and knowledge of the desired reconstruction phase already during the CT scan, a deviation of the later actually used from the previously planned Reconstruction phase would lead to a strong loss of image quality.

Since the heart movement is a non-uniform quasiperiodic motion, not all phase segments are equivalent in terms of phase-correlated image reconstruction. For example, in the systolic phase where the heart is performing the ejection movement, the heart has a high rate of movement. In the diastolic phase, on the other hand, after relaxation, the heart is in a short, sedentary resting phase. Thus, in most cases, the diastolic phase, which is between 60% and 80% of an RR cycle of an ECG signal, is the optimal reconstruction phase. However, as the heart rate increases, so does the systolic phase, which is between 20% and 40% of a RR cycle, for artifact-free reconstruction. Both the uncertainty about a reconstruction in the systolic or diastolic phase and a slight patient-specific deviation within the typical phase relationships (systole, diastole) do not allow a clear statement about a general optimal reconstruction phase.

For this reason, attempts have been made, on the one hand, to iteratively approximate optimal image quality by repeated reconstructions at different phase points. On the other hand, image-based approaches exist to determine the optimal reconstruction phase both globally (over the entire recording time) and locally (at specific times). Both possibilities, however, are very time-consuming and resource-intensive due to the many necessary image reconstructions. In addition, it is not possible the optimal Reconstructing phase already to be determined during the CT scan, since for the previously known solutions the reconstructed images and thus the complete raw data are needed.

An object of the present invention is to enable automatic determination of this optimal reconstruction phase in view of reduction of motion artifacts.

This object is achieved by a method for displaying a quasi-periodically moving object with the aid of an image recording system, in particular a computer tomograph, using a phase-correlated reconstruction technique in which an optimal image reconstruction phase is determined using a motion function of the object.

In addition, this object is achieved by an image acquisition system, in particular computed tomography, for

Representation of a quasiperiodisch moving object using a phase-correlated

Reconstruction technique solved by a device for

Determining the optimal image reconstruction phase using a motion function of the object is characterized.

Finally, the invention is also achieved by using the motion function of an object to determine an optimal image reconstruction phase in a method of displaying a quasi-periodically moving object using a phase-correlated reconstruction technique and a computer program for solves an above-mentioned image pickup system comprising computer program instructions for carrying out the above-mentioned method when the computer program is executed on a computer.

In this case, a quasi-periodic signal is understood as meaning a signal with a recurring signal profile, the period of which, however, being variable and not necessarily constant. In this sense, however, a quasi-periodic signal can also be strictly periodic.

The present invention modifies the current standard reconstruction technique for computed tomography (CT) -based imaging of moving objects. To obtain images free of motion artifacts, a phase-correlated reconstruction technique is used. The present invention describes a computer tomograph with a fully automatic, computer-aided determination of the optimal reconstruction phase to minimize the

Motion artifacts. For this purpose, the phase section of the quasi-periodic motion function of the moving object is determined from a movement function of the object, in particular the kymogram function, in which a minimal movement takes place. The size of the considered phase section is determined by the system parameters of the imaging modality used. With the invention, the achievable image quality is significantly increased and thus increases the diagnostic value of the images.

In contrast to the existing solution approaches, the method described relies on the direct analysis of the actual movement function of the object, ie for example, the actual heart movement function, not a similarity calculation or third-party data. In other words, the determination of the optimal reconstruction phase is not based on data describing the object movement only indirectly or indirectly, but on the basis of direct and immediate motion data. The determination of the optimal reconstruction phase takes place exclusively in the raw data space, ie in particular using CT raw data. The image quality is thus improved without the need for steps beyond the already performed image reconstruction. In particular, no calculations in the image space, ie based on already reconstructed CT images, are required. This significantly reduces the total time required for the actual CT scan, resulting in a noticeably improved CT scan.

The core of the invention is a method that allows to individually calculate the optimal phase for an image reconstruction, in real time for each individual period of the movement, in real time. As a result, the time required for the image reconstruction or the image quality can be improved. For this purpose, the kymogram signal is calculated from the raw data at fixed time intervals, which reflects the time course of the center of mass, from which in turn its speed and finally the optimal reconstruction phase, namely that with minimal movements, can be calculated. In one embodiment of the invention, that of the

Kymogrammsignal resulting minimal object velocity detected and thus determines the optimal reconstruction phase. In a further embodiment of the invention These individual and period-dependent results are compared with averaged curves (templates) to reduce the influence of the noise Σ: u.

In addition, an implementation for real-time computation, which is the prerequisite for optimal tube current modulation, is described. This results in a fast raw data-based method for determining the patient-specific optimal reconstruction phase while reducing the radiation load for the patient compared to image-based methods.

The terms kymogram, kymogram function and kymogram signal are used synonymously in this description. A kymogram is understood as movement information as described in the following publications:

Marc Kachelrieß, Dirk-Alexander Sennst, Wolfgang Maximoser, and Willi A. Kalender: "Kymogram detection and kymogram-correlated image reconstruction from subsecond spiral computed tomography scans of the heart". Medical Physics, 29 (7): 1489-1503,

W. A. Kalender and M. Kachelrieß: "Computed tomography with object-related motion artifact reduction and extraction of the object motion information kymogram." European Patent Application No. 99111708.6.

The content of these two publications is hereby incorporated in full in the present description. Details of the kymogram function and its detection are known from the above publications, so that in the context of this description need not be discussed in more detail.

Further advantageous embodiments of the invention are specified in the subclaims and are described below in

Described in connection with an embodiment of the invention, which is explained in more detail with reference to the following drawings. These show:

1 is a schematic CT system with its essential for the present invention components,

Fig. 2 shows the ECG signal and the corresponding

Kymogrammfunktion (above) and the RR cycle of the ECG signal and the corresponding (nodular

Phase p (t) (below),

3 movement function of an example patient,

4 shows a cardio-CT image created using a conventional reconstruction technique and a cardio-CT image created using a phase-correlated reconstruction technique.

The present invention describes the detection of the optimal motion phase for phase-correlated image reconstruction using the calculated cardiac motion function, the kymogram. This allows a motion artifact-free image reconstruction in the optimal reconstruction phase. A patient-specific adaptation of the reconstruction phase is no longer necessary. The invention will be described below using the example of a cardio-CT system 1. This consists essentially of imaging elements 2 (X-ray tube, detectors, etc.) and an associated control component 3 and a Bildrekoristruktionskomponente 4, which creates CT images using the raw data obtained from the imaging elements 2, see. FIG. 1. The steps essential to the invention described below are predominantly realized by the control component 3.

In this case, the control component 3 comprises at least one data processing unit with a number of function modules explained in greater detail below, wherein each function module is designed to perform a specific function or a number of specific functions according to the described method. The function modules can be hardware modules or software modules. In other words, as far as the data processing unit is concerned, the invention can be implemented either in the form of computer hardware or in the form of computer software or in a combination of hardware and software. As far as the invention is implemented in the form of software, the functions described below are realized by computer program instructions when the computer program is executed on a computer. The computer may be, for example, a standard personal computer or a dedicated medical workstation. The computer program instructions are implemented in a manner known per se in any programming language and can be provided to the data processing unit in any form, for example in the form of data packets which are transmitted via a computer network or in the form of a diskette, a CD-ROM or the like. ROM or a computer program product stored on a different medium.

The image reconstruction component 4 also comprises at least one data processing unit and a computer software designed correspondingly for image reconstruction. Such an image reconstruction component is known in the art. In particular, it may be a computer unit installed remotely from the imaging elements 2.

The control component 3 comprises a first functional module 5, which is designed to provide the kymogram function of the examined heart. For this purpose, the first functional module 5 with the imaging elements 2 of the CT

Systems 1 connected via a data line 101. The kymogram is calculated by the first function module 5 directly from the CT raw data. In contrast to an ECG ₇ in which this is only indirectly the case, with the kymogram the actual movements of the heart are detected. On the known structure and the operation of this first functional module 5 of the control component 3 will not be discussed further below. The kymogram signal is provided by the first functional module 5 to a second functional module 6 of the control component 3 or transmitted to it via a data line 103. This second functional module 6 then performs the individual steps described in detail below.

The two-dimensional kymogram signal

reflects the movement of the center of mass (COM point) of the x-rayed layer of the heart in the x- and y-direction, respectively. Due to the pumping motion of the heart, the COM point varies with time and therefore reflects the heart movement. FIG. 2 (top) shows an ECG signal 7 and the corresponding kymogram function 8.

Since the kymogram signal is a quasi-periodic signal, only the phase motion function r _c (p) of a motion cycle is for the invention

(Movement period) with respect to the movement phase p of interest. For this purpose, the individual motion cycles with the period length T are considered within the continuous time signal r _c (t) with te [t _S tart # t _En d]. The movement phase p represents the relative time within the period, ie the time within a period _tp € [0, T [normalized to the period T:

P

A representative, patient-specific

Phase motion function is preferably obtained by averaging the motion function r _c (t) over the recording duration for a selected motion phase p:

In other words, a motion function of the COM points is generated which is representative of a motion cycle of the COM

Heart. Here, p represents 1 2 (fc) € [0, [modular phase 9 the signal is corresponding to the motion phase p at the time te [ts _tart, _End] t and exemplified in Figure (below) is depicted.:

The variable heart rate is in this case by 1 / T _n = I / (t _{n +} t i _n) and the synchronization points by t _{n +} i, t _n _n t _{+ 1>} _n t indicated. The modular phase refers to an arbitrary synchronization signal, which reflects the quasi-periodic object movement.

For example, the kymogram function itself can be used as the synchronization signal. In particular, in the cardio-CT an additional ECG 10 is used, which is connected via a data line 102 to the control component 3 of the CT system 1. The RR cycle 11 of such an ECG signal is shown in Fig. 2 (below).

A phase motion function 12 of an example patient r _c (p) for p {t) e [0, 1 [with 100 samples is shown in FIG. The gray scale coding allows an assignment to the corresponding ECG signal in FIG. 2 (bottom). The light gray measuring points shown in the lower left area of FIG. 3 corresponding to the systolic phase (p (t) = 20% -40%) and the darker measuring points shown in the middle areas correspond to the diastolic phase (p (t) = 60%). 80%). The velocity of the heart with respect to the motion phase can be determined with the aid of the distance of two adjacent phase-based COM points r _c {p) and r _c (p + Δp) to one another. At a constant phase-based sampling rate Δp, two adjacent COM points at higher speeds have a greater distance \ r _c (p + Δp) -r _c (p) | on than at lower speeds.

The optimal reconstruction _phase p _opt is determined by solving a minimization problem in the second functional module 6 of the control component 3. In this case, the distance of the phase COM point r _c (p) with respect. of the phase point p to the neighboring phase points r _c (p + pi) is minimized by a relevant range 2 -p _w :

The relevant region 2 -p _w (width of the time window for the phase-correlated image reconstruction) is defined by the relative time resolution in the specific example of a phase-correlated image reconstruction in the cardio-CT. This is from the patient's heart rate used

Reconstruction algorithm (including single-segment, multi-segment) and the system parameters of the CT system 1 dependent. For example, the window width would increase at higher heart rates and approximately halve when using a multi-segment reconstruction algorithm with, for example, two windows. After the optimal reconstruction _phase p _{opt is determined} in the second functional _module 6 of the control component 3, a corresponding activation of the component 4 for image reconstruction takes place via a data line 106. In particular, a selection element

Image reconstruction component 4 transmit control signals that enable the selection element to select only those CT raw data for image reconstruction, which include a minimum possible object movement. The necessary measurement data for image reconstruction receives the corresponding component 4 via the data line 107 from the imaging element 2. A comparison of a conventional

Reconstructed cardiac CT image 13 and an image 14 made with a phase-correlated reconstruction technique, as shown in FIG. 4, illustrates the improved image quality.

In a further embodiment of the invention, in the second functional module, an online (i.e. real-time) capable calculation of the optimal reconstruction phase for a

Tube current modulation (TCM) performed. For this purpose, the kymogram function r _c (fc) is already calculated during the CT scan. However, the phase motion function r _c (p) can then not be calculated by averaging over the entire recording duration of the kymogram function r _c (t), but instead results in a phase movement function dependent on the time of recording:

'begin

Using the above-mentioned minimization method thus results for the optimal reconstruction phase P _o p _t (t) is a time dependency. From this time dependence, the timing of the optimal reconstruction phase within the next cycle of motion can be predicted. This is done using appropriately adapted calculation algorithms in the second functional module 6 of the control component. In other words, from already determined values, the optimum reconstruction phase for the next heart movement cycle can be predicted. Depending on this, then an activation of the imaging elements 2 of the CT system 1 via a

Data line 104, in particular the X-ray tube, by the second functional module 6 such that in motion-rich phases that provide low image contributions, the tube current as a function of the cardiac phase p (fc) designed and thus the dose burden is reduced.

In a further embodiment of the invention, a correlation calculation is carried out with a template curve (sample movement pattern). For this purpose, previously determined patient-specific

Phase motion functions r _c (p) averaged over a certain number of patients to a representative template curve. The template curve is then stored in the second functional module 6 or provided to the second functional module 6 for use, for example, from an external database via a data line 105, which also for data transport of result or control data of the control component 3 to an external receiving station (not shown) can be.

During a CT scan, the second functional module uses the minimization method described above to make the optimal reconstruction phases global, within one Movement cycle, or locally, for example, within the systolic or diastolic phase, determined in the Tetnplatekurve. By means of a correlation calculation of the patient-specific phase movement function r _c (p) or r _c (p, t) with the template curve also carried out by the second function module, both the optimal reconstruction phase and characteristic movement phases, such as the midsystolic and diastolic phases, can be determined become. In other words, a comparison of measured data with pattern data takes place so that an optimal reconstruction phase already determined in advance can be found very quickly. This correlation calculation is applicable to both online (ie, real-time) and standard off-line calculation.

Template curves can be provided in particular for different heart rate ranges. Thus, the heart rate dependency of the optimal reconstruction phase can be met in a simple manner.

All in the description, the following claims and the drawings illustrated features may be essential to the invention both individually and in any combination.

LIST OF REFERENCE NUMBERS

1 CT system 2 imaging elements

3 control component

4 image reconstruction component

5 first functional module 6 second functional module

7 ECG signal

8 kymogram

9 modular phase 10 ECG

11 RR cycle

12 phase motion function

13 conventional CT imaging

14 enhanced CT-Bid 101-107 data lines

Claims

claims

A method of displaying a quasi-periodically moving object using an image acquisition system (2), in particular a computed tomography scanner, using a phase-correlated reconstruction technique, characterized in that an optimal image reconstruction phase is determined using a motion function (8) of the object.

2. The method according to claim 1, characterized in that it is a kymogram in the movement function.

3. Method according to claim 1 or 2, characterized by a control of at least individual elements (2) of the

Imaging system (1) according to the determined optimal image reconstruction phase.

4. The method according to any one of claims 1 to 3, characterized by a control of at least individual elements of a

Image reconstruction component (4) corresponding to the determined optimum image reconstruction phase.

5. Image recording system (1), in particular computed tomography, for displaying a quasiperiodisch moving

An object using a phase-correlated reconstruction technique, characterized by a device (5) for determining the optimal image reconstruction phase using a motion function (8) of the object.

6. Use of the motion function (8) of an object for determining an optimal image reconstruction phase in a method for representing a quasi-periodic moving object using a phase-correlated reconstruction technique.

A computer program for an image capturing system (1) according to claim 5, comprising computer program instructions for carrying out the method according to any one of claims 1 to 4 when the computer program is executed on a computer (3).