UW CVES: PROJECTS


		LABORATORY STUDIES AT THE CENTER FOR VIDEOENDOSCOPIC SURGERY In 1997, the CVES completed an installation of a dedicated experimental surgical laboratory. Four dedicated videoendoscopic surgical stations, augmented with full ethernet and AV connectivity to the operating room and conference room provide an optimal forum for educational and research projects. A number of research programs are planned for this lab. FORCE/TORQUE SIGNATURES IN MINIMALLY INVASIVE SURGERY To address the problem of training surgeons in skills for performing advanced laparoscopic surgical procedures, a six axis force sensing laparoscopic instrument was developed to measure forces applied during surgical maneuvers. The grasper handle has been designed with several fully interchangeable standard surgical tools and allows an entire operation to be performed with complete measurement of all forces applied in the six axes. A specially designed user interface which included real-time force-torque was simultaneously recorded on videotape with the endoscopic view of the procedure using picture in picture video technology. Each operative procedure was then divided into specific tasks and analyzed frame by frame. The instrumented grasper and its hardware were integrated with a laptop using specialized software for data acquisition. The seven channels of data (Fx, Fy, Fz, Tx, Ty, Tz, Fg) were sampled at a frequency of 30 Hz. (Fig. 2) and the data analyzed with two types of algorithms: K-means vector quantization (VQ), and the Hidden Markov Model (HMM). The experimental data sets were processed by the VQ algorithm in order /torque signatures. These signatures were then related to the states of the HMM. Each state in HMM was defined by identifiable surgical manuvers (e.g. grasping & pulling, spreading & pushing, etc.) and HMM validation of each operative procedure verified by manual frame by frame analysis of operative video footage. Preliminary studies with the grasper have yielded the following data. Comparing two experienced and two inexperienced laparoscopic surgeons force and torque profiles have revealed that force used to perform an operation was 138% and torque 130% greater in the inexperienced surgeons. Furthermore, the time taken to complete a laparoscopic cholecystectomy on a porcine model was 270% greater in the non-surgeon group. The non-surgeon group was also noted to spend more time in an "idle" position without performing specific tasks toward performing the operation. EFFECT OF CONVENTIONAL LAPAROSCOPIC FORCE/TORQUES ON TISSUE DAMAGE To measure whether the increase in torque and force generated translates into biologic damage, tissue specimens were obtained and reviewed by a veterinary pathologist in an attempt to grade this damage. The results of these studies are still being analyzed. Over the next year, experienced laparoscopic and intern-level, trainee surgeons will perform cholecystectomies and Nissen fundoplications on a porcine model in order to generate Markov profiles for comparison. As was done previously, picture in picture video monitoring will allow for assessment of surgical skill. Additional tissue will also be harvested and fixed by the pathologist in an attempt to quantify the tissue damage incurred by these different measured forces and torques. These projects will serve to generate further information as a means of providing a more characteristic picture of force/torque signatures in experienced compared to inexperienced surgeons. Ultimately, it is hoped that aberrations of force signatures discovered using this technology will give a picture as to which steps of an operation require additional training in order for an individual to become surgically proficient.		Educational Studies Clinical Studies


This page last updated 11/20/98 Comments to CVESWebmaster@u.washington.edu ©1996, 1997 University of Washington Department of Surgery Center for Videoendoscopic Surgery Home Page University of Washington Department of Surgery Home Page University of Washington Home Page