Prediction of open urinary tract in laparoscopic partial nephrectomy by virtual resection plane visualization

Background: The purpose of this study is presenting a method to predict the presence of an open urinary tract and the position of the opening in laparoscopic partial nephrectomy from three dimensional (3D) computed tomography (CT) images by using novel image segmentation and visualization techniques. Methods. From CT images of patients who underwent laparoscopic partial nephrectomy, 3D regions of the kidney, urinary tract, and tumor were segmented. For each patient, multiple virtual resection planes of the kidney with different surgical margins (1 mm to 5 mm, every 1 mm) were generated and the presence of an open urinary tract and the position of the opening were predicted from the images. Results: We compared the predictions with actual operations in 5 cases by using recorded video of the operations and operative notes. In terms of the presence of an open urinary tract, agreement of the predictions and the intraoperative results was obtained in all patients. The expected positions of the openings were close to those in the actual operations. Conclusions: We have developed a method to virtually visualize the resection plane of laparoscopic partial nephrectomy. Image segmentation methods used in this study were precise and effective. The comparison indicated that our method accurately predicted the presence of an open urinary tract and the position of the opening and provided useful preoperative information.