The CycleGAN framework allows for unsupervised image-to-image translation of unpaired data. In a scenario of surgical training on a physical surgical simulator, this method can be used to transform endoscopic images of phantoms into images which more closely resemble the intra-operative appearance of the same surgical target structure. This can be viewed as a novel augmented reality approach, which we coinedHyperrealismin previous work. In this use case, it is of paramount importance to display objects like needles, sutures or instruments consistent in both domains while altering the style to a more tissue-like appearance. Segmentation of these objects would allow for a direct transfer, however, contouring of these, partly tiny and thin foreground objects is cumbersome and perhaps inaccurate. Instead, we propose to use landmark detection on the points when sutures pass into the tissue. This objective is directly incorporated into a CycleGAN framework by treating the performance of pre-trained detector models as an additional optimization goal. We show that a task defined on these sparse landmark labels improves consistency of synthesis by the generator network in both domains. Comparing a baseline CycleGAN architecture to our proposed extension (DetCycleGAN), mean precision (PPV) improved by$+61.32$, mean sensitivity (TPR) by$+37.91$, and mean$F_1$score by$+0.4743$. Furthermore, it could be shown that by dataset fusion, generated intra-operative images can be leveraged as additional training data for the detection network itself.

The mitral valve, one of the four valves in the human heart, controls the bloodflow between the left atrium and ventricle and may suffer from various pathologies. Malfunctioning valves can be treated by reconstructive surgeries, which have to be carefully planned and evaluated. While current research focuses on the modeling and segmentation of the valve, we base our work on existing segmentations of patient-specific mitral valves, that are also time-resolved ( 3D+t) over the cardiac cycle. The interpretation of the data can be ambiguous, due to the complex surface of the valve and multiple time steps. We therefore propose a software prototype to analyze such 3D+t data, by extracting pathophysiological parameters and presenting them via dimensionally reduced visualizations. For this, we rely on an existing algorithm to unroll the convoluted valve surface towards a flattened 2D representation. In this paper, we show that the 3D+t data can be transferred to 3D or 2D representations in a way that allows the domain expert to faithfully grasp important aspects of the cardiac cycle. In this course, we not only consider common pathophysiological parameters, but also introduce new observations that are derived from landmarks within the segmentation model. Our analysis techniques were developed in collaboration with domain experts and a survey showed that the insights have the potential to support mitral valve diagnosis and the comparison of the pre- and post-operative condition of a patient.

Echocardiography (cardiac ultrasound) is today the predominant technique for quantitative assessment of cardiac function and valvular heart lesions. Segmentation of cardiac structures is required to determine many important diagnostic parameters. As the heart is a moving organ, reliable information can be obtained only from three-dimensional (3-D) data over time (3-D + time = 4-D). Due to their size, the resulting four-dimensional (4-D) data sets are not reasonably accessible to simple manual segmentation methods. Automatic segmentation often yields unsatisfactory results in a clinical environment, especially for ultrasonic images. We describe a semiautomated segmentation algorithm (ROPES) that is able to greatly reduce the time necessary for user interaction and its application to extract various parameters from 4-D echocardiographic data. After searching for candidate contour points, which have to fulfill a multiscale edge criterion, the candidates are connected by minimizing a cost function to line segments that then are connected to form a closed contour. The contour is automatically checked for plausibility. If necessary, two correction methods that can also be used interactively are applied (fitting of other line segments into the contour and searching for additional candidates with a relaxed criterion). The method is validated using in vivo transesophageal echocardiographic data sets.