Kendall Maxwell (polanddrain63)

A fully connected layer was trained on the intermediate feature representation to classify instrument-tissue interaction. RESULTS The perception study revealed that acoustic feedback has potential to improve the perception during MIS and to serve as a basis for further automated analysis. The proposed classification pipeline yielded excellent performance for four types of instrument-tissue interaction (muscle, fascia, liver and fatty tissue) and achieved top-1 accuracies of up to 89.9%. Moreover, our model is able to distinguish electrosurgical operation modes with an overall classification accuracy of 86.40%. CONCLUSION Our proof-of-principle indicates great application potential for guidance systems in MIS, such as controlled tissue resection. Supported by a pilot perception study with surgeons, we believe that utilizing audio signals as an additional information channel has great potential to improve the surgical performance and to partly compensate the loss of haptic feedback.PURPOSE For laparoscopic ablation to be successful, accurate placement of the needle to the tumor is essential. Laparoscopic ultrasound is an essential tool to guide needle placement, but the ultrasound image is generally presented separately from the laparoscopic image. We aim to evaluate an augmented reality (AR) system which combines laparoscopic ultrasound image, laparoscope video, and the needle trajectory in a unified view. METHODS We created a tissue phantom made of gelatin. Artificial tumors represented by plastic spheres were secured in the gelatin at various depths. The top point of the sphere surface was our target, and its 3D coordinates were known. The participants were invited to perform needle placement with and without AR guidance. Once the participant reported that the needle tip had reached the target, the needle tip location was recorded and compared to the ground truth location of the target, and the difference was the target localization error (TLE). The time of the needle placement was also recorded. We further tested the technical feasibility of the AR system in vivo on a 40-kg swine. RESULTS The AR guidance system was evaluated by two experienced surgeons and two surgical fellows. The users performed needle placement on a total of 26 targets, 13 with AR and 13 without (i.e., the conventional approach). The average TLE for the conventional and the AR approaches was 14.9 mm and 11.1 mm, respectively. The average needle placement time needed for the conventional and AR approaches was 59.4 s and 22.9 s, respectively. For the animal study, ultrasound image and needle trajectory were successfully fused with the laparoscopic video in real time and presented on a single screen for the surgeons. CONCLUSION By providing projected needle trajectory, we believe our AR system can assist the surgeon with more efficient and precise needle placement.PURPOSE Sustained delivery of regenerative retinal therapies by robotic systems requires intra-operative tracking of the retinal fundus. We propose a supervised deep convolutional neural network to densely predict semantic segmentation and optical flow of the retina as mutually supportive tasks, implicitly inpainting retinal flow information missing due to occlusion by surgical tools. METHODS As manual annotation of optical flow is infeasible, we propose a flexible algorithm for generation of large synthetic training datasets on the basis of given intra-operative retinal images. We evaluate optical flow estimation by tracking a grid and sparsely annotated ground truth points on a benchmark of challenging real intra-operative clips obtained from an extensive internally acquired dataset encompassing representative vitreoretinal surgical cases. Wnt-C59 RESULTS The U-Net-based network trained on the synthetic dataset is shown to generalise well to the benchmark of real surgical videos. When used to track retinal points of interest, our flow estimation outperforms variational bas