Accurate morphological information on aortic valve cusps is critical in treatment planning. Image segmentation is necessary to acquire this information, but manual segmentation is tedious and time consuming. In this paper, we propose a fully automatic aortic valve cusps segmentation method from CT images by combining two deep neural networks, spatial configuration-Net for detecting anatomical landmarks and U-Net for segmentation of aortic valve components. A total of 258 CT volumes of end systolic and end diastolic phases, which include cases with and without severe calcifications, were collected and manually annotated for each aortic valve component. The collected CT volumes were split 6:2:2 for the training, validation and test steps, and our method was evaluated by five-fold cross validation. The segmentation was successful for all CT volumes with 69.26 s as mean processing time. For the segmentation results of the aortic root, the right-coronary cusp, the left-coronary cusp and the non-coronary cusp, mean Dice Coefficient were 0.95, 0.70, 0.69, and 0.67, respectively. There were strong correlations between measurement values automatically calculated based on the annotations and those based on the segmentation results. The results suggest that our method can be used to automatically obtain measurement values for aortic valve morphology.

Objective . Accurate extraction of mitral valve shape from clinical tomographic images acquired in patients has proven useful for planning surgical and interventional mitral valve treatments. However, manual extraction of the mitral valve shape is laborious, and the existing automatic extraction methods have not been sufficiently accurate. In this paper, we propose a fully automated method of extracting mitral valve shape from computed tomography (CT) images for the all phases of the cardiac cycle. Approach . This method extracts the mitral valve shape based on DenseNet using both the original CT image and the existence probability maps of the mitral valve area inferred by U-Net as input. A total of 1585 CT images from 204 patients with various cardiac diseases including mitral regurgitation were collected and manually annotated for mitral valve region. The proposed method was trained and evaluated by 10-fold cross validation using the collected data and was compared with the method without the existence probability maps. Main results . The mean error of shape extraction error in the proposed method is 0.88 mm, which is an improvement of 0.32 mm compared with the method without the existence probability maps. Significance . We present a novel fully automatic mitral valve extraction method from input to output for all phases of 4D CT images. We suggest that the accuracy of mitral valve shape extraction is improved by using existence probability maps.

In recent years, transcatheter edge-to-edge repair (TEER) has been widely adopted as an effective treatment for mitral regurgitation (MR). The aim of this study is to develop a personalized in silico model to predict the effect of edge-to-edge repair in advance to the procedure for each individual patient. For this purpose, we propose a combination of a valve deformation model for computing the mitral valve (MV) orifice area (MVOA) and a lumped parameter model for the hemodynamics, specifically mitral regurgitation volume (RVol). Although we cannot obtain detailed information on the three-dimensional flow field near the mitral valve, we can rapidly simulate the important medical parameters for the clinical decision support. In the present method, we construct the patient-specific pre-operative models by using the parameter optimization and then simulate the postoperative state by applying the additional clipping condition. The computed preclip MVOAs show good agreement with the clinical measurements, and the correlation coefficient takes 0.998. In addition, the MR grade in terms of RVol also has good correlation with the grade by ground truth MVOA. Finally, we try to investigate the applicability for the predicting the postclip state. The simulated valve shapes clearly show the well-known double orifice and the improvement of the MVOA, compared with the preclip state. Similarly, we confirmed the improved reverse flow and MR grade in terms of RVol. A total computational time is approximately 8 h by using general-purpose PC. These results obviously indicate that the present in silico model has good capability for the assessment of edge-to-edge repair.

Background and objective Tricuspid regurgitation (TR) is one of the most common forms of valvular heart diseases. The morphological information of the tricuspid valve annulus (TVA) is critical in treatment planning for TR. It is necessary to extract the TVA from medical images to obtain that information, however this task is difficult and time-consuming to perform manually. In this paper, we propose a method to automatically extract and measure the TVA from computed tomography (CT) images. Methods: Our proposed method coarsely crops CT images to the region surrounding the tricuspid valve based on the right atrium and the right ventricle regions. The cropped CT images are input to a stacked hourglass network with loss function integrating the mean squared error loss, the focal loss and the shape-aware weighted Hausdorff distance loss to extract 36 landmarks on the TVA. The extraction accuracy of TVA landmarks was evaluated by five-fold cross validation using 120 CT images with manually annotated TVA landmarks. In addition, measurements of TVA morphology based on automatically extracted TVA and those based on manually annotated TVA were calculated and compared using the same measurement algorithm which provides a means to automatically generate seven measurements based on TVA landmarks. Results: Our proposed method extracted TVA inside the right heart in all CT images without any processing interruption. The mean processing time was 27.09 ± 8.65 s, and the Chamfer distance and Hausdorff distance were 2.07 ± 0.53 and 4.09 ± 1.29, respectively. The mean absolute error between the measurements based on automatically extracted TVA and those based on manually annotated TVA was less than 4 mm, which is less than the typical device size interval for surgical prosthetic valve rings in current use, for measurement items related to distance. For all seven measurement items, significant correlations (r = 0.51–0.99, p < 0.0071) were shown between the measurements based on automatically extracted TVA and those based on manually annotated TVA. Conclusions: Our proposed method was able to automatically extract and measure the TVA. This method is expected to reduce the time and effort required by physicians in treatment planning for TR.

Aims : Preoperative observation of computed tomography (CT) cross sections corresponding to standard transesophageal echocardiographic (TEE) views is useful for treatments using TEE. However, manually locating those CT cross sections from the CT volume is time consuming. This study aimed to develop a fully automatic method to locate those CT cross sections corresponding to standard TEE views. Methods : Developed method crops CT volume based on the heart segmentation model if input is a chest CT. CT cross sections corresponding to standard TEE views are coarsely located from cropped chest CT or cardiac CT based on the esophagus segmentation model and the plane localization model. The coarse cross sections are refined based on the anatomical landmark detection model. Accuracies of located CT cross sections were evaluated by five-fold cross validation with comparison to manually set cross sections, using 110 CT volumes. The results were also compared to the top and the middle axial section of the heart as baselines. Results : The mean of rotation error, probe position error, landmark error and structural similarity index between the predicted and manually set cross sections were 12.52 ± 8.27 degree, 13.90 ± 10.62 mm, 3.95 ± 3.48 mm and 0.64 ± 0.15, respectively, and these errors were significantly smaller than the baselines. The mean processing time was 41.34 ± 22.75 s. Conclusions : Our developed method can provide CT cross sections corresponding to standard TEE views without user operations, resulting in faster treatment planning with less effort and without inter user variability in treatments using TEE.