Processing of medical image data of the heart for deformation analysis

Abstract

A method for generating segmented, masked 4D image data (M-BD) of the heart is described. In this method, a 4D image, preferably an angiographic image, of a patient's heart (O) is provided, which contains 4D image data (4D-BD) of the patient's heart (O). Subsequently, first segmented 4D image data (SG1-BD) are generated based on the 4D image data (4D-BD), segmenting the cardiac wall of one ventricle. Furthermore, second segmented 4D image data (SG2-BD) are generated based on the first segmented 4D image data (SG1-BD), segmenting the epicardium (EPD) and the endocardium (EKD) of the ventricle. Finally, segmented, masked 4D image data (M-BD) are generated based on the second segmented 4D image data (SG2-BD), masking the interior of the ventricle. Furthermore, a cardiac deformation analysis method is described. An image data generation device (50) is also described.Furthermore, a medical imaging system is described.

Description

[0001] The invention relates to a method for generating segmented, masked 4D image data of the heart. Furthermore, the invention relates to a cardiac deformation analysis method. The invention also relates to an image data generation device. Furthermore, the invention relates to a computed tomography system.

[0002] Modern imaging techniques are used to generate image data that serves to visualize an imaged object under examination. The image data or image data sets obtained in this way can also be used for further applications or analyses. The present invention relates to a further application or analysis of image data or image data sets obtained in this way.

[0003] Ischemic heart disease is one of the most common causes of death worldwide. Reduced blood flow due to plaque deposits that narrow the blood vessels leads to insufficient blood supply to the heart muscle and limited function. One way to diagnose the disease is by measuring the volume of blood pumped during a heart contraction. This evaluation measures cardiac ejection fraction and is part of Cardiac Functional Analysis (CFA). The disadvantage of this method is that it can only detect the disease in later stages because it is based on an indirect manifestation.

[0004] Cardiac strain analysis (CSA), on the other hand, allows the detection of small, regionally limited functional cardiac abnormalities even in seemingly asymptomatic patients. Cardiac strain analysis measures the cardiac motion that leads to thickening and thinning of the myocardium and thus to strain. Therefore, cardiac strain analysis is suitable as a complementary method for identifying and localizing myocardial dysfunction.

[0005] Non-invasive imaging modalities are most commonly used to detect cardiac deformations throughout the cardiac cycle. In clinical settings, cardiac deformation analysis is primarily performed using echocardiography. However, echocardiography-based cardiac deformation analysis is known to be sensitive to image acquisition quality. Therefore, it is prone to errors, and the results are not always reliable.

[0006] Another imaging technique currently gaining popularity in the field of cardiac deformation analysis is cardiac magnetic resonance imaging (CMRI). It offers better image quality in terms of noise reduction and tissue differentiation, enabling precise CMRI-based CSA. However, CMRI is not widely available and involves a time-consuming acquisition procedure. CMRI is described in M.M. Lamacie et al., "Quantifying myocardial deformation by deformable registration-based analysis of cine MRI: Validation with tagged CMR," European Radiology, vol. 29, no. 7, pp. 3658–3668, 2019, doi: 10.1007 / s00330-019-06019-9.

[0007] In cardiac imaging, computed tomography (CT) is primarily used for coronary angiography—a method for visualizing coronary arteries using contrast media. Performing cardiac deformation analysis using cardiac CT imaging is an area of ​​active research. Previous approaches to cardiac deformation analysis using CT scans were met with moderate success. They generally failed due to the inability of motion estimation algorithms to handle low tissue contrast in CT images. Therefore, CT cardiac deformation analysis is not yet performed in clinical practice.

[0008] Existing approaches for performing cardiac deformation analysis using CT images are inaccurate and therefore insufficient for clinical application. Some attempts to apply motion estimation approaches from other modalities to CT have met with moderate success. One such approach is described in F. Ammon et al., "CT-derived left ventricular global strain: a direct comparison with speckle-tracking echocardiography," The International Journal of Cardio Imaging, vol. 35, no. 9, pp. 1701-1707, 2019, doi: 10.1007 / s10554-019-01596-8 and in SJ Buss et al., "Quantitative Analysis of Left Ventricular Strain Using Cardiac Computed Tomography", European Journal of Radiology, Vol. 83, No. 3, e123-e130, 2014, doi: 10.1016 / j.ejrad.2013.11.026.

[0009] Another approach, which estimates myocardial motion based on a model fitted to cardiac CT images from different time points, appears to be the most promising in the state of the art, although the approach may not be fully automated, is limited to the left ventricle, and suffers from certain inaccuracies. This approach is described in Z. Peled et al., "Automated 4-dimensional regional myocardial strain assessment using cardiac computed tomography," The International Journal of Cardio Imaging, vol. 36, no. 1, pp. 149–159, 2020, doi: 10.1007 / s10554-019-01696-5, and in Y. Lamash et al., "Strain Analysis From 4-D Cardiac CT Image Data," IEEE Transactions on Biomedical Engineering, vol. 62, No. 2, pp. 511-521, 2015, doi: 10.1109 / TBME.2014.2359244. described.

[0010] The documents US20200193603A1, “Automated segmentation utilizing fully convolutional networks”, US20220328195A1, “System and method for heart strain determination”, and US20080015428A1, “Motion-guided segmentation for cine dense images” are known from the state of the art.

[0011] It is therefore a task to adapt 4D CT imaging of the heart so that it is suitable for cardiac deformation analysis.

[0012] This object is achieved by a method for generating medical image data for a heart deformation analysis according to patent claim 1, a heart deformation analysis method according to patent claim 11, an image data generation device according to patent claim 12 and by a computed tomography system according to patent claim 13.

[0013] In the method according to the invention for generating segmented, masked 4D image data of the heart, a 4D image data acquisition, preferably image data from an angiography acquisition, particularly preferably a coronary angiography acquisition, of a patient's heart is provided, preferably as a CT image acquisition, which comprises 3D or 4D image data of the patient's heart. Coronary angiography is an angiography of the coronary arteries and thus a special form of X-ray examination in which the coronary arteries are imaged. In angiography, contrast agents are typically used for imaging to visualize blood vessels.

[0014] The method according to the invention for generating segmented, masked 4D image data of the heart as well as the heart deformation analysis method according to the invention described below is designed in particular as a computer-implemented method.

[0015] In addition to coronary angiography, all other 4D data acquisition techniques that can delineate the muscle of a ventricle from the blood within it should also be included. Therefore, such an acquisition may also include a high-resolution CT scan without contrast medium. The acquisition may also include a transesophageal 4D ultrasound scan. Alternatively, the 4D data acquisition may also include an MRI scan.

[0016] Based on the 4D image data, initial segmented 4D image data are generated, segmenting the heart wall of a ventricle, preferably the left ventricle. The application of the inventive method to the left ventricle is preferred, since the condition of the left ventricle has the greatest influence on the functioning of the systemic circulation. The right ventricle has a much thinner wall than the left ventricle. The function of the blood-pulmonary circulation, which is influenced by the right ventricle, is generally not as clinically relevant as the function of the systemic circulation. However, at this point, imaging of the right ventricle is expressly included, albeit less prominently.

[0017] A method for heart chamber segmentation is described in Y. Zheng et al., "Four-chamber heart modeling and automatic segmentation for 3-D cardiac CT volumes using marginal space learning and steerable features," IEEE transactions on medical imaging, vol. 27, no. 11, pp. 1668-1681, 2008, doi: 10.1109 / TMI.2008.2004421. The aforementioned document is incorporated into the present patent application by reference.

[0018] 4D image data refers to time-resolved three-dimensional image data. The heart wall is formed by the myocardium, the heart muscle tissue. Segmentation is performed in such a way that the heart wall of the ventricle, preferably the left ventricle, is delineated from its surroundings. As will be explained in more detail later, measures are preferably taken to highlight the heart wall of the ventricle, preferably the left ventricle, from the image areas surrounding the heart wall, so that any movement of the heart wall of the ventricle, particularly the left ventricle, particularly a contraction, dilation, or expansion of the heart wall, can be tracked more precisely and clearly.

[0019] Based on the first segmented 4D image data, second segmented 4D image data are generated, segmenting the epicardium and endocardium of the ventricle. The epicardium forms the outermost layer or outer lining of the heart. It is fused to the myocardium. The endocardium forms the innermost layer of the heart, which, as a smooth endocardium, covers the entire inner surface of the heart. The second segmented 4D image data preferably includes highlighting the endocardium and epicardium relative to their surroundings, allowing movement of the endocardium and epicardium to be tracked more precisely and clearly.

[0020] Based on the second segmented 4D image data, segmented, masked 4D image data is generated, obscuring the interior of the ventricle, preferably the left ventricle. The interior of a ventricle, especially the left ventricle, contains the valve muscles of the heart valves, which also move when the ventricle deforms. Movement of the valve muscles can be mistakenly interpreted as movement of the heart wall of the ventricle in question, which would distort the deformation analysis.

[0021] Advantageously, the ventricle, preferably the left ventricle, and its outlines are made more clearly visible and more clearly delineated from the surrounding area, enabling, among other things, more precise surface mapping. Furthermore, by increasing the contrast of the cardiac structures, improved motion tracking is possible.

[0022] By enhancing the images of adjacent cardiac phases, intensity-based registration between images is better focused on the actual region of interest for LV myocardial strain analysis or cardiac deformation analysis. The registration algorithms provide deformation fields from which intrinsic myocardial trajectories can be derived.

[0023] Intensity-based registration algorithms utilize smoothness constraints to create realistic-looking motion estimates. Movement around the inspected object can mislead the registration process and lead to inaccurate motion estimates for the object. This effect can be eliminated by masking out misleading motion / objects in the image.

[0024] When the ventricle, preferably the left ventricle, contracts, an increase in density occurs within the heart muscle, which is indicated by an increase in brightness or contrast in this area. Due to the better utilization of the range of possible brightness values, a contraction can be quantitatively recorded more accurately. Conversely, expansion or dilation can also be detected by detecting a reduction in density. In addition, the distances between the epicardium and the endocardium also change during contraction and dilation, which is also easier to detect using the segmentation described above.

[0025] Twists that occur during contraction can also be detected if necessary.

[0026] The method according to the invention for generating segmented, masked 4D image data of the heart is particularly advantageously applicable to low-contrast imaging methods on the basis of which a cardiac function analysis, in particular a cardiac deformation analysis, is to be carried out.

[0027] In the cardiac deformation analysis method according to the invention, the method according to the invention for generating segmented, masked 4D image data of the heart is first performed. Next, a deformation field is determined based on the segmented, masked 4D image data generated by the method. For this purpose, the dynamic 4D image data is registered with a reference image.

[0028] The deformation field then comprises displacement vectors from pixels of an object image to pixels of a reference image, which are determined by an elastic registration process part.

[0029] Based on the determined deformation field, an intrinsic myocardial movement trajectory is determined. Based on the determined intrinsic myocardial movement trajectory, a myocardial deformation analysis of the ventricle, preferably the left ventricle, is performed.

[0030] Due to the improved image data underlying the cardiac deformation analysis method, a more precise analysis of the heart's deformation is possible, which can improve the validity and accuracy of a prognosis of a patient's heart health.

[0031] The image data generation device according to the invention has an input interface for receiving 4D image data from the patient's heart. The image data generation device according to the invention also includes a first segmentation unit for generating first segmented 4D image data based on the 4D image data, wherein the heart wall of a ventricle, preferably the left ventricle, is segmented.

[0032] The image data generating device according to the invention also has a second segmentation unit for generating second segmented 4D image data based on the first segmented 4D image data, wherein the epicardium and the endocardium of the ventricle, preferably of the left ventricle, are segmented.

[0033] Furthermore, the image data generation device according to the invention comprises a mask unit for generating segmented, masked 4D image data based on the second segmented 4D image data, wherein the interior of the ventricle, preferably the left ventricle, is masked. The image data generation device according to the invention shares the advantages of the method according to the invention for generating segmented, masked 4D image data of the heart.

[0034] The medical imaging system according to the invention comprises a scanning unit for acquiring measurement data, preferably projection data, from a patient and a control device for controlling the scanning unit and for generating image data based on the measurement data, preferably the projection data. An image data generation device according to the invention is also part of the medical imaging system according to the invention. The imaging system according to the invention shares the advantages of the method according to the invention for generating segmented, masked 4D image data of the heart.

[0035] In particular, the features and advantages described in connection with the methods according to the invention can also be designed as corresponding subunits of the medical imaging system according to the invention or of the computer program product according to the invention.

[0036] Conversely, the features and advantages described in connection with the medical imaging system according to the invention and the computer program product according to the invention can also be designed as corresponding method steps of the methods according to the invention.

[0037] The medical imaging system preferably comprises a computed tomography system. The computed tomography system according to the invention has a scanning unit for acquiring projection data from a patient and a control device for controlling the scanning unit and for generating image data based on the projection data. An image data generation device according to the invention is also part of the computed tomography system according to the invention. The computed tomography system according to the invention shares the advantages of the method according to the invention for generating segmented, masked 4D image data of the heart.

[0038] Alternatively, the medical imaging system may also include an ultrasound imaging system with transesophageal 4D ultrasound. Alternatively, the 4D data acquisition may also include an MRI scan.

[0039] The computer program product according to the invention has program code sections with which all steps of the method according to the invention for generating image data based on the projection data or the heart deformation analysis method are carried out when the program is executed in a control device of a medical imaging system, preferably a computed tomography system.

[0040] A largely software-based implementation has the advantage that existing medical imaging systems, preferably computer tomography systems, or their control devices can be easily retrofitted by a software update in order to operate in the manner according to the invention.

[0041] A large portion of the aforementioned components of the image data generation device according to the invention can be implemented entirely or partially in the form of software modules in a processor of a corresponding computing system, e.g., a control device of a medical imaging system or a computer used to control such a system. A largely software-based implementation has the advantage that even previously used computing systems can be easily upgraded to operate in the manner according to the invention via a software update.In this respect, the object is also achieved by a corresponding computer program product with a computer program that can be loaded directly into a computer system, with program sections for executing the steps of the inventive method for generating segmented, masked image data of the heart based on the projection data or the steps of the cardiac deformation analysis method when the program is executed in the computer system. Such a computer program product may, in addition to the computer program, optionally include additional components, such as documentation, and / or additional components, including hardware components, such as hardware keys (dongles, etc.), for using the software.

[0042] A computer-readable medium, e.g., a memory stick, a hard disk, or another portable or permanently installed data storage device, can be used for transport to the computer system or control device and / or for storage on or in the computer system or control device, on which the program sections of the computer program that can be read and executed by a computer system are stored. For this purpose, the computer system can, for example, have one or more cooperating microprocessors or the like.

[0043] The dependent claims and the following description each contain particularly advantageous embodiments and developments of the invention. In particular, the claims of one claim category can also be developed analogously to the dependent claims of another claim category. Furthermore, within the scope of the invention, the various features of different embodiments and claims can also be combined to form new embodiments.

[0044] In a variant of the method according to the invention for generating segmented, masked 4D image data of the heart, the step of generating first segmented 4D image data comprises generating a mask of the wall of the left ventricle. The mask of the wall of the left ventricle marks the region relevant for a deformation analysis of the heart. Making this region more visible facilitates registration of the heart walls and their surfaces, in particular the epicardium and the endocardium. Thus, the accuracy and reliability of a deformation analysis of the left ventricle can be improved based on registration of the heart walls and surfaces of the thus improved 4D image data. The mask of the wall of the left ventricle is preferably generated by applying a four-chamber segmentation algorithm to the 4D image data.Advantageously, the left ventricle can be segmented from the other heart chambers using the four-chamber segmentation algorithm.

[0045] In one variant of the inventive method for generating segmented, masked 4D image data of the heart, a high-contrast mask image is generated. The voxels within the mask whose gray values ​​or brightness values ​​deviate from a statistical mean by a maximum of a predetermined value are determined. The gray values ​​are mapped to high-contrast gray values ​​that lie between the gray value 0 and the sum of the statistical mean and the predetermined value. Advantageously, the scale of the brightness values ​​in the mask region is better utilized to better visualize textures and contrasts.

[0046] In a preferred variant of the method according to the invention for generating segmented, masked 4D image data of the heart, the predetermined value is two standard deviations from the mean. With a normal distribution, 95 percent of the values ​​of the distribution lie within the range of values, which is limited by two standard deviations around the mean. Advantageously, the range of possible brightness values ​​is defined such that most of the values ​​of the voxels located within the mask area lie within it.

[0047] In a particularly preferred embodiment of the method according to the invention for generating segmented, masked 4D image data of the heart, the first segmented 4D image data are generated by combining the high-contrast mask image with the weighted 4D image data. Advantageously, the image data for the surroundings of the mask region are retained, and the contrast in the mask region is improved by adding the higher-contrast 4D image data within the mask region.

[0048] In a preferred variant of the method according to the invention for generating segmented, masked 4D image data of the heart, the weighted combination is performed at a predetermined percentage. Advantageously, the original image information is also contained in the segmented image data. In particular, the voxels outside the wall of the left ventricle are unchanged.

[0049] In a preferred embodiment of the method according to the invention for generating segmented, masked 4D image data of the heart, the percentage for the high-contrast mask image relative to the segmented 4D image data is between 20 and 40 percent, preferably approximately 30 percent. This means that the high-contrast mask image is adopted at 20 to 40 percent and the original image at 60 to 80 percent for the segmented, masked 4D image data. At a value of 30 percent, the high-contrast mask image is weighted at 30 percent and the original 4D image data at 70 percent. Advantageously, the original image information is retained in the segmented, masked 4D image data of the heart with a predominant proportion.

[0050] Preferably, the second segmented 4D image data is generated by highlighting the contrast of the epicardium and endocardium of the left ventricle. Advantageously, the surfaces of the left ventricle are clearly highlighted, facilitating the registration process and enabling more precise tracking of the movement of the left ventricular surfaces.

[0051] When generating second segmented 4D image data based on the first segmented 4D image data, the brightness values ​​of the region segmented as epicardium and endocardium are preferably adjusted to a mean value plus twice the standard deviation of this mean value. Advantageously, the surfaces of the left ventricle are made particularly visible.

[0052] Preferably, an interface enhancement factor is set to preserve texture information on the surface of the left ventricle. This interface enhancement factor is used to determine the percentage of voxel values ​​within the surface that are set to the aforementioned mean plus twice the standard deviation from this mean, and the percentage of original values ​​for voxels in the surface that are retained in the resulting image. The gray values ​​in the surface of the left ventricle can be multiplied by an additional factor greater than 1. This makes the endocardium and epicardium particularly visible in the resulting image and facilitates intensity-based registration.

[0053] The step of generating segmented, masked 4D image data involves creating and inverting a mask for the intraventricular blood lumen, which encompasses the endocardial tissue. The mask for the intraventricular blood lumen is inverted in such a way that the entire blood-filled volume or blood lumen is covered. Advantageously, the portion of the left ventricle to be imaged is segmented or separated from the intraventricular blood lumen. As already mentioned, the interior of a ventricle contains the valve muscles of the heart valves, which also move when the ventricle deforms. Movement of the valve muscles can be mistakenly interpreted as movement of the ventricular wall, which would distort the deformation analysis. Therefore, covering the interior of the ventricle is advantageous for cardiac deformation analysis.

[0054] The invention is explained in more detail below with reference to exemplary embodiments and the accompanying figures. They show: Fig. 1 is a flowchart illustrating a method for generating segmented, masked 4D image data of the heart according to an embodiment of the invention, Fig. 2 a sequence of images showing step 1.I and step 1.II of the Fig. 1 illustrated method according to an embodiment of the invention, Fig. 3 a diagram which also shows step 1.I and step 1.II of the Fig. 1 illustrated method according to an embodiment of the invention, Fig. 4 a picture showing the steps 1.III and 1.IV of the Fig. 1 illustrated method according to an embodiment of the invention, Fig. 5 a schematic representation of an image data generation device according to an embodiment of the invention, Fig. 6 is a flowchart illustrating a heart deformation analysis method according to an embodiment of the invention, Fig. 7 is a schematic representation of a heart deformation analysis device according to an embodiment of the invention, Fig. 8 a schematic representation of a computed tomography system according to an embodiment of the invention.

[0055] In Fig. 1 is a flowchart illustrating a method for generating segmented, masked 4D image data M-BD of the heart according to an embodiment of the invention.

[0056] In step 1.I, a coronary angiography image of the heart of a patient O is provided, which image comprises 4D image data 4D-BD of the heart of the patient O.

[0057] Based on the 4D image data 4D-BD, first segmented 4D image data SG1-BD are generated in step 1.II, whereby the heart wall of the left ventricle of the heart of patient O is segmented.

[0058] In step 1.III, second segmented 4D image data SG2-BD are generated based on the first segmented 4D image data SG1-BD, whereby the epicardium and the endocardium of the left ventricle are segmented.

[0059] Finally, in step 1.IV, segmented, masked 4D image data M-BD are generated based on the second segmented 4D image data SG2-BD, whereby the inner region of the left ventricle is masked.

[0060] In Fig. 2 is a sequence of images showing step 1.I and step 1.II of the Fig. 1 according to an embodiment of the invention. The image sequence in Fig. 2 comprises three subimages. The left subimage illustrates image data BD of a perspective view of the heart with a left ventricle LV. A middle subimage shows a long-axis view SA of the left ventricle, and a right subimage shows a short-axis view DA of the left ventricle. Segmentation lines that delimit the ventricular muscle internally and externally are displayed in the middle and right subimages. The outer dashed segmentation lines ASL delimit the muscle tissue externally toward the pericardium, while the inner dotted segmentation lines ISL delimit the muscle tissue internally toward the endocardium.

[0061] In Fig. 3 shows a diagram showing step 1.I and also step 1.II of the Fig. 1 according to an embodiment of the invention. In step 1.I, the previously described image of a patient's heart is taken, with 4D image data 4D-BD being acquired from the heart of patient O.

[0062] In step 1.II, in addition to the segmentation of the left ventricle, which is Fig. 2, also a modified image representation, whereby the contrast of the representation of the segmented area, i.e. the left ventricle, is increased.

[0063] In sub-step 1.IIa, a mask is created that covers the wall of the left ventricle. Subsequently, in sub-step 1.IIb, the brightness values ​​or gray values ​​of the voxels encompassed by the mask are statistically analyzed. A mean of the brightness values ​​or gray values, measured in Hounsfield Units (HU), is determined. Furthermore, in step 1.IIb, a standard deviation from the mean is determined. With a normal distribution, the brightness values ​​of 95% of all voxels lie within a value range that is characterized by the mean (MW) plus / minus twice the standard deviation (STAW) from the mean (MW) as the boundary values.

[0064] In sub-step 1.IIc, the brightness values ​​of the voxels within the mask are restricted to this value range. In particular, brightness values ​​outside the specified value range are set to the boundary values, i.e., the aforementioned mean value (MW) plus / minus twice the standard deviation (STAW).

[0065] In sub-step 1.IId, the brightness values ​​of the voxels located in the mask area are redistributed to an expanded value interval EWI, which ranges from 0 to a value formed by the sum of the mean value MW and twice the standard deviation STAW. In this way, 4D image data 4D-BDK with increased contrast is generated in the mask area. In sub-step 1.IIe, the original 4D image data 4D-BD and the 4D image data 4D-BDK with increased contrast generated in sub-step 1.IId are combined or overlaid with a weighting factor, generating combined 4D image data 4D-BDKK with increased contrast. The weighting factor specifies the percentage of the 4D image data 4D-BDK with increased contrast in the mask area and the percentage of the original 4D image data 4D-BD in the mask area that should be combined.The voxels of the original 4D image data 4D-BD outside the mask area are kept unchanged.

[0066] In Fig. 4 is a picture showing steps 1.III and 1.IV of the Fig. 1 according to an embodiment of the invention. In step 1.III, the edge regions of the ventricle are provided with a type of surface mask. The mask can be determined by a morphological dilation operation. The edge regions are determined starting from a point inside the mask or the 4D image data 4D-BDK with increased contrast in the mask region generated in step 1.II. Subsequently, the brightness values ​​of the surface mask are increased to the maximum value of the previously generated value range, i.e., to the mean value plus twice the standard deviation STAW. To retain the texture information in the surface mask, the original brightness values ​​of the pixels in the area of ​​the surface mask can also be combined with the increased brightness values ​​in a weighted manner. In this way, the endocardium EKD and the epicardium EPD are illuminated. In step 1.IV, segmented, masked 4D image data (M-BD) are generated by shadowing the interior of the left ventricle. This means that the interior of the left ventricle, which is brightly displayed due to the contrast agent administration and the associated increased X-ray absorption, is inverted with respect to its X-ray attenuation, i.e., displayed darkly. In Fig. Figure 5 shows a schematic representation of an image data generation device 50 according to an embodiment of the invention. The image data generation device 50 includes an input interface 51 configured to receive 4D image data 4D-BD from the patient's heart.

[0067] Part of the image data generation device 50 is also a first segmentation unit 52, which is configured to generate first segmented 4D image data SG1-BD based on the 4D image data 4D-BD, wherein the heart wall of the left ventricle is segmented.

[0068] The image data generation device 50 also comprises a second segmentation unit 53 which is configured to generate second segmented 4D image data SG2-BD based on the first segmented 4D image data SG1-BD, wherein the epicardium and the endocardium of the left ventricle are segmented.

[0069] Furthermore, the image data generation device 50 has a mask unit 54 which is configured to generate segmented, masked 4D image data M-BD based on the second segmented 4D image data SG2-BD, wherein the inner region of the left ventricle is concealed.

[0070] In Fig. 6, a flowchart 600 is shown illustrating a cardiac deformation analysis method according to an embodiment of the invention.

[0071] In step 6.I, the Fig. 1 to 4, the method for generating segmented, masked 4D image data M-BD of the heart is carried out according to an embodiment of the invention.

[0072] In step 6.II, a deformation field DF is determined based on the segmented, masked 4D image data M-BD generated by the method. In step 6.III, an intrinsic myocardial motion trajectory T is determined based on the determined deformation field.

[0073] In step 6.IV, a myocardial deformation analysis of the left ventricle is performed based on the determined intrinsic myocardial movement trajectory and a result AE of the deformation analysis is output.

[0074] In Fig. 7 shows a schematic representation of a heart deformation analysis device 70 according to an embodiment of the invention.

[0075] The cardiac deformation analysis device 70 has an input interface 71 for receiving segmented, masked 4D image data M-BD of the heart. The segmented, masked 4D image data M-BD of the heart are received by an image data generation device 50, as shown in Fig. 5. Also part of the cardiac deformation analysis device 70 is a field determination unit 72, which is configured to determine a deformation field DF based on the segmented, masked 4D image data M-BD generated during the method.

[0076] Part of the cardiac deformation analysis device 70 is also a trajectory determination unit 73, which is configured to determine an intrinsic myocardial movement trajectory T on the basis of the determined deformation field DF.

[0077] Part of the cardiac deformation analysis device 70 is also an analysis unit 74, which is configured to perform a deformation analysis of the left ventricle on the basis of the determined intrinsic myocardial movement trajectory T and to output a result AE of the deformation analysis.

[0078] In Fig. 8 is a schematic representation of a computed tomography system 80 according to an embodiment of the invention.

[0079] The computed tomography system 80 has a control device 81 and a scanning unit 82 controlled by the control device 81. The control device 81 has a control unit 83 for generating control data SD and a control interface 84 for transmitting control signals SS, which are generated on the basis of the control data SD, to the scanning unit 82. Part of the control device 81 is also an input interface 85, which is configured to receive projection data PD from the scanning unit 82. The projection data PD are transmitted to a reconstruction unit 86, which is also part of the control device 81 and is configured to reconstruct 4D image data 4D-BD of the heart of a patient (not shown) lying in the scanning unit 82. Part of the control device 81 is also the Fig. 5, which is configured to generate mask image data M-BD based on the 4D image data 4D-BD of the heart. The control device 81 also includes the Fig. 7 illustrates a cardiac deformation analysis device 70 which is configured to perform a deformation analysis based on the generated segmented, masked 4D image data M-BD and to output result data AE of the analysis.

[0080] Finally, it should be noted once again that the detailed methods and structures described above are exemplary embodiments, and that the basic principle can be varied widely by those skilled in the art without departing from the scope of the invention, as defined by the claims. It should be noted that the features of all exemplary embodiments or developments disclosed in the figures can be used in any combination.

[0081] For the sake of completeness, it should also be noted that the use of the indefinite articles "ein" and "eine" does not exclude the possibility that the relevant characteristics may be present multiple times. Likewise, the term "unit" does not exclude the possibility that this unit consists of multiple components, which may also be spatially distributed. Regardless of the grammatical gender of a particular term, this includes persons with male, female, or other gender identities.

Claims

[1] Method for generating segmented, masked 4D image data (M-BD) of the heart, comprising the steps: - Providing a 4D image of the heart of a patient (O), which has 4D image data (4D-BD) of the heart of the patient (O), - Generating first segmented 4D image data (SG1-BD) based on the 4D image data (4D-BD), whereby the heart wall of a ventricle of the heart is segmented, - Generating second segmented 4D image data (SG2-BD) based on the first segmented 4D image data (SG1-BD), whereby the epicardium (EPD) and the endocardium (EKD) of the ventricle are segmented, - generating segmented, masked 4D image data (M-BD) based on the second segmented 4D image data (SG2-BD), wherein the inner region of the ventricle is obscured, wherein the step of generating segmented, masked 4D image data (M-BD) comprises generating and inverting an intraventricular mask. [2] The method of claim 1, wherein the ventricle comprises the left ventricle (LV). [3] The method of claim 1 or 2, wherein the step of generating first segmented 4D image data (SG1-BD) comprises generating a mask of the wall of the ventricle (LV). [4] The method of claim 3, wherein the mask is generated by applying a four-chamber segmentation algorithm to the 4D image data (4D-BD). [5] Method according to claim 3 or 4, wherein a high-contrast mask image is generated, wherein the voxels within the mask are determined whose gray values ​​deviate from a statistical mean value (MW) by a maximum of a predetermined value and the gray values ​​are mapped to high-contrast gray values ​​which lie between the gray value 0 and the sum of the statistical mean value (MW) and the predetermined value. [6] Method according to claim 5, wherein the predetermined value is two standard deviations (STAW) from the mean value (MW). [7] Method according to claim 5 or 6, wherein the first segmented 4D image data (SG1-BD) are generated by combining the high-contrast mask image with the 4D image data (4D-BD) in a weighted manner. [8] A method according to claim 7, wherein the weighted combination is carried out at a predetermined percentage. [9] The method of claim 8, wherein the percentage is between 20 and 40 percent for the high-contrast mask image relative to the first segmented 4D image data (SGB1-BD), preferably about 30 percent. [10] Method according to one of the preceding claims, wherein - the second segmented 4D image data (SG2-BD) are generated by highlighting the contrast of the epicardium (EPD) and the endocardium (EKD) of the ventricle (LV). [11] A cardiac deformation analysis method comprising the steps of: - Carrying out the method according to one of claims 1 to 10, - Determining a deformation field (DF) based on the segmented, masked 4D image data (M-BD) generated by the method, - Determination of an intrinsic myocardial movement trajectory (T) based on the determined deformation field (DF), - Performing a myocardial deformation analysis of the left ventricle (LV) based on the determined intrinsic myocardial movement trajectory (TG). [12] Image data generating device (50), comprising: - an input interface (51) for receiving 4D image data (4D-BD) from the heart of a patient (O), - a first segmentation unit (52) for generating first segmented 4D image data (SG1-BD) on the basis of the 4D image data (4D-BD), wherein the heart wall of a ventricle of the heart is segmented, - a second segmentation unit (53) for generating second segmented 4D image data (SG2-BD) on the basis of the first segmented 4D image data (SG1-BD), wherein the epicardium (EPD) and the endocardium (EKD) of the ventricle are segmented, - a mask unit (54) for generating segmented, masked 4D image data (M-BD) on the basis of the second segmented 4D image data (SG2-BD), wherein the inner region of the ventricle is concealed, wherein the generation of segmented, masked 4D image data (M-BD) comprises generating and inverting an intraventricular mask. [13] Medical imaging system, preferably a computed tomography system (80), comprising: - a scanning unit (82) for acquiring measurement data (PD) from a patient (O), - a control device (81) for controlling the scanning unit (82) and for generating 4D image data (4D-BD) on the basis of the measurement data (PD), - an image data generating device (50) according to claim 12. [14] Computer program product comprising a computer program which is directly loadable into a memory of a control device (81) of a medical imaging system, with program code sections to carry out all steps of a method according to one of claims 1 to 11 when the program is executed in the control device (81). [15] Computer-readable medium on which program sections executable by a computer unit are stored in order to carry out the steps of a method according to one of claims 1 to 11 when the program sections are executed by the computer unit.

Images