A Novel Approach for Semiautomated Three-Dimensional Quantification of Mitral Regurgitant Volume Reflects a More Physiologic Approach to Mitral Regurgitation.

BACKGROUND: Quantification of mitral regurgitation (MR) by echocardiography is integral to assessing lesion severity and entails the integration of multiple Doppler-based parameters. These methods are founded primarily upon the principle of proximal isovelocity surface area (PISA), a two-dimensional (2D) method known to involve several assumptions regarding MR jet characteristics. The authors analyzed the results of a semiautomated method of three-dimensional (3D)-based regurgitant volume (RVol) estimation that accounts for jet behavior throughout the cardiac cycle and compared it with conventional 2D PISA methods for MR quantification. METHODS: A total of 50 patients referred for transesophageal echocardiography for evaluation of primary (n = 25) and secondary (n = 25) MR were included for analysis. Three-dimensional full-volume color data sets were acquired, along with standard 2D methods for PISA calculation. A 3D semiautomated MR flow quantification algorithm was applied offline to calculate 3D RVol, with simultaneous temporal curves generated from the 3D data set. Three-dimensional RVol was compared with 2D RVol. Three-dimensional vena contracta area was also performed in all cases. RESULTS: There was a modest correlation between 2D RVol and 3D RVol (r = 0.60). The semiautomated 3D approach resulted in significantly lower values of RVol compared with 2D PISA. Real-time and dynamic flow curve patterns were used for integral estimates of 3D RVol over the cardiac cycle, with a distinct bimodal pattern in functional MR and a brief and solitary peak in primary MR. CONCLUSIONS: Using a semiautomated 3D software for the quantification of MR allows the simultaneous calculation of 3D RVol with an automated generation of dynamic flow curves characteristic of the underlying MR mechanism. The present flow curve pattern results highlight well-known differences between MR flow dynamics in degenerative MR compared with functional MR.

A conventional-to-spectral CT image translation augmentation workflow for robust contrast injection-independent organ segmentation.

PURPOSE: In computed tomography (CT) cardiovascular imaging, the numerous contrast injection protocols used to enhance structures make it difficult to gather training datasets for deep learning applications supporting diverse protocols. Moreover, creating annotations on noncontrast scans is extremely tedious. Recently, spectral CT's virtual-noncontrast images (VNC) have been used as data augmentation to train segmentation networks performing on enhanced and true-noncontrast (TNC) scans alike, while improving results on protocols absent of their training dataset. However, spectral data are not widely available, making it difficult to gather specific datasets for each task. As a solution, we present a data augmentation workflow based on a trained image translation network, to bring spectral-like augmentation to any conventional CT dataset. METHOD: The conventional CT-to-spectral image translation network (HUSpectNet) was first trained to generate VNC from conventional housnfied units images (HU), using an unannotated spectral dataset of 1830 patients. It was then tested on a second dataset of 300 spectral CT scans by comparing VNC generated through deep learning (VNCDL ) to their true counterparts. To illustrate and compare our workflow's efficiency with true spectral augmentation, HUSpectNet was applied to a third dataset of 112 spectral scans to generate VNCDL along HU and VNC images. Three different three-dimensional (3D) networks (U-Net, X-Net, and U-Net++) were trained for multilabel heart segmentation, following four augmentation strategies. As baselines, trainings were performed on contrasted images without (HUonly) and with conventional gray-values augmentation (HUaug). Then, the same networks were trained using a proportion of contrasted and VNC/VNCDL images (TrueSpec/GenSpec). Each training strategy applied to each architecture was evaluated using Dice coefficients on a fourth multicentric multivendor single-energy CT dataset of 121 patients, including different contrast injection protocols and unenhanced scans. The U-Net++ results were further explored with distance metrics on every label. RESULTS: Tested on 300 full scans, our HUSpectNet translation network shows a mean absolute error of 6.70 ± 2.83 HU between VNCDL and VNC, while peak signal-to-noise ratio reaches 43.89 dB. GenSpec and TrueSpec show very close results regardless of the protocol and used architecture: mean Dice coefficients (DSCmean ) are equal with a margin of 0.006, ranging from 0.879 to 0.938. Their performances significantly increase on TNC scans (p-values < 0.017 for all architectures) compared to HUonly and HUaug, with DSCmean of 0.448/0.770/0.879/0.885 for HUonly/HUaug/TrueSpec/GenSpec using the U-Net++ architecture. Significant improvements are also noted for all architectures on chest-abdominal-pelvic scans (p-values < 0.007) compared to HUonly and for pulmonary embolism scans (p-values < 0.039) compared to HUaug. Using U-Net++, DSCmean reaches 0.892/0.901/0.903 for HUonly/TrueSpec/GenSpec on pulmonary embolism scans and 0.872/0.896/0.896 for HUonly/TrueSpec/GenSpec on chest-abdominal-pelvic scans. CONCLUSION: Using the proposed workflow, we trained versatile heart segmentation networks on a dataset of conventional enhanced CT scans, providing robust predictions on both enhanced scans with different contrast injection protocols and TNC scans. The performances obtained were not significantly inferior to training the model on a genuine spectral CT dataset, regardless of the architecture implemented. Using a general-purpose conventional-to-spectral CT translation network as data augmentation could therefore contribute to reducing data collection and annotation requirements for machine learning-based CT studies, while extending their range of application.

Spectral augmentation for heart chambers segmentation on conventional contrasted and unenhanced CT scans: an in-depth study.

PURPOSE: Recently, machine learning has outperformed established tools for automated segmentation in medical imaging. However, segmentation of cardiac chambers still proves challenging due to the variety of contrast agent injection protocols used in clinical practice, inducing disparities of contrast between cavities. Hence, training a generalist network requires large training datasets representative of these protocols. Furthermore, segmentation on unenhanced CT scans is further hindered by the challenge of obtaining ground truths from these images. Newly available spectral CT scanners allow innovative image reconstructions such as virtual non-contrast (VNC) imaging, mimicking non-contrasted conventional CT studies from a contrasted scan. Recent publications have demonstrated that networks can be trained using VNC to segment contrasted and unenhanced conventional CT scans to reduce annotated data requirements and the need for annotations on unenhanced scans. We propose an extensive evaluation of this statement. METHOD: We undertake multiple trainings of a 3D multi-label heart segmentation network with (HU-VNC) and without (HUonly) VNC as augmentation, using decreasing training dataset sizes (114, 76, 57, 38, 29, 19 patients). At each step, both networks are tested on a multi-vendor, multi-centric dataset of 122 patients, including different protocols: pulmonary embolism (PE), chest-abdomen-pelvis (CAP), heart CT angiography (CTA) and true non-contrast scans (TNC). An in-depth comparison of resulting Dice coefficients and distance metrics is performed for the networks trained on the largest dataset. RESULTS: HU-VNC-trained on 57 patients significantly outperforms HUonly trained on 114 regarding CAP and TNC scans (mean Dice coefficients of 0.881/0.835 and 0.882/0.416, respectively). When trained on the largest dataset, significant improvements in all labels are noted for TNC and CAP scans (mean Dice coefficient of 0.882/0.416 and 0.891/0.835, respectively). CONCLUSION: Adding VNC images as training augmentation allows the network to perform on unenhanced scans and improves segmentations on other imaging protocols, while using a reduced training dataset.

Characterizing breast lesions through robust multimodal data fusion using independent diffuse optical and x-ray breast imaging.

To enable tissue function-based tumor diagnosis over the large number of existing digital mammography systems worldwide, we propose a cost-effective and robust approach to incorporate tomographic optical tissue characterization with separately acquired digital mammograms. Using a flexible contour-based registration algorithm, we were able to incorporate an independently measured two-dimensional x-ray mammogram as structural priors in a joint optical/x-ray image reconstruction, resulting in improved spatial details in the optical images and robust optical property estimation. We validated this approach with a retrospective clinical study of 67 patients, including 30 malignant and 37 benign cases, and demonstrated that the proposed approach can help to distinguish malignant from solid benign lesions and fibroglandular tissues, with a performance comparable to the approach using spatially coregistered optical/x-ray measurements.

Segmentation of thrombus in abdominal aortic aneurysms from CTA with nonparametric statistical grey level appearance modeling.

This paper presents a new method for deformable model-based segmentation of lumen and thrombus in abdominal aortic aneurysms from computed tomography (CT) angiography (CTA) scans. First the lumen is segmented based on two positions indicated by the user, and subsequently the resulting surface is used to initialize the automated thrombus segmentation method. For the lumen, the image-derived deformation term is based on a simple grey level model (two thresholds). For the more complex problem of thrombus segmentation, a grey level modeling approach with a nonparametric pattern classification technique is used, namely k-nearest neighbors. The intensity profile sampled along the surface normal is used as classification feature. Manual segmentations are used for training the classifier: samples are collected inside, outside, and at the given boundary positions. The deformation is steered by the most likely class corresponding to the intensity profile at each vertex on the surface. A parameter optimization study is conducted, followed by experiments to assess the overall segmentation quality and the robustness of results against variation in user input. Results obtained in a study of 17 patients show that the agreement with respect to manual segmentations is comparable to previous values reported in the literature, with considerable less user interaction.

Efficient model-based quantification of left ventricular function in 3-D echocardiography.

Quantitative functional analysis of the left ventricle plays a very important role in the diagnosis of heart diseases. While in standard two-dimensional echocardiography this quantification is limited to rather crude volume estimation, three-dimensional (3-D) echocardiography not only significantly improves its accuracy but also makes it possible to derive valuable additional information, like various wall-motion measurements. In this paper, we present a new efficient method for the functional evaluation of the left ventricle from 3-D echographic sequences. It comprises a segmentation step that is based on the integration of 3-D deformable surfaces and a four-dimensional statistical heart motion model. The segmentation results in an accurate 3-D + time left ventricle discrete representation. Functional descriptors like local wall-motion indexes are automatically derived from this representation. The method has been successfully tested both on electrocardiography-gated and real-time 3-D data. It has proven to be fast, accurate, and robust.