Search

Computational dentistry uses computerized methods and mathematical models for dental image analysis. One of the fundamental problems in computational dentistry is accurate tooth instance segmentation in high-resolution mesh data of intra-oral scans (IOS). This paper presents a new computational model based on deep neural networks, called Mask-MCNet, for end-to-end learning of tooth instance segmentation in 3D point cloud data of IOS. The proposed Mask-MCNet localizes each tooth instance by predicting its 3D bounding box and simultaneously segments the points that belong to each individual tooth instance. The proposed model processes the input raw 3D point cloud in its original spatial resolution without employing a voxelization or down-sampling technique. Such a characteristic preserves the finely detailed context in data like fine curvatures in the border between adjacent teeth and leads to a highly accurate segmentation as required for clinical practice (e.g. orthodontic planning). The experiments show that the Mask-MCNet outperforms state-of-the-art models by achieving 98% Intersection over Union (IoU) score on tooth instance segmentation which is very close to human expert performance.

Over the past years High Definition (HD) endoscopy has become a crucial tool for the early detection of esophageal cancer. The high resolution offers specialist physicians high-quality visual information, enabling them to identify dysplastic tissue leading to Early Adenocarcinoma (EAC). The detection and removal of these early types of cancer drastically increases the survival chances of the patient. However, even for an experienced specialist it remains an arduous task to identify the patterns associated with early cancer. Therefore, a computer-aided detection system that supports the physician seems highly attractive. We present a novel algorithm for automatic detection of early cancerous tissue in HD endoscopic images. The algorithm computes local color- and texture features based on the original and on the Gabor-filtered image. We explore the spectral characteristics of the image regions that contain early cancer and we design appropriate filters based on this analysis. The features are classified by a trained Support Vector Machine (SVM) after which additional post-processing techniques are applied in order to annotate the image region containing early cancer. For 7 patients, we compare 32 annotations made by the algorithm with the corresponding delineations made by an expert gastroenterologist. Of 38 lesions indicated independently by the gastroenterologist, the system detects 36 of those lesions with a recall of 0.95 and a precision of 0.75.

Autostereoscopic visualization in clinical image-guided interventions and therapy poses constrains on the maximal latency in the visualization chain. In order to use visual feedback in the hand-eye coordinated loop, the latency of the visualization chain should be less than approximately 250–300 ms. This is a challenging constraint for interactive autostereoscopic multi-view volume rendering of large datasets. The various building blocks of such a visualization chain and their latency aspects are explored in this paper. Two complementary strategies to improve the latency of the entire chain are introduced and examined: lowering the resolution of the rendered views and reducing the amount of rendered views by smartly interpolating the missing views at receiver side. These strategies are balanced while optimizing the latency and image quality using a fuzzy logic approach. Furthermore, the optimal view resolution for a multi-view autostereoscopic lenticular display has been determined by investigating the lenticular lattice in the frequency domain. The quantitative aspects of the latency of the proposed building blocks and the resulting image quality have been measured.