A valid method of laparoscopic simulation training and competence assessment.

BACKGROUND: The purpose of our study was to evaluate the construct validity of laparoscopic technical performance measures and the face validity of three laparoscopic simulations. MATERIALS AND METHODS: Subjects (N = 27) of varying levels of surgical experience performed three laparoscopic simulations, representing appendectomy (LA), cholecystectomy (LC), and inguinal hemiorrhaphy (LH). Five laparoscopic surgeons, blinded to the identity of the subjects, rated the subjects on procedural competence on a binary scale and in four skills categories on a 5-point scale: clinical judgment, dexterity, serial/simultaneous complexity, and spatial orientation. Using a task-specific checklist, non-clinical staff assessed the technical errors. The level of surgical experience was correlated with the ratings, the technical errors, and the time for each procedure. Subject responses to a survey regarding the utility of the inanimate models were evaluated. RESULTS: Years of experience directly correlated with the skills ratings (all P < 0.001) and with the competence ratings across the three procedures (P < 0.01). Experience inversely correlated with the time for each procedure (P < 0.01) and the technical error total across the three models (P < 0.05). Nearly all subjects agreed that the corresponding procedures were well represented by the simulations (LA 96%, LC 96%, LH 100%). CONCLUSION: The laparoscopic simulations demonstrated both face and construct validity. Regardless of the level of surgical experience, the subjects found the models to be suitable representations of actual laparoscopic procedures. Task speed improved with surgical experience. More importantly, the quality of performance increased with experience, as shown by the improvement in the skills assessments by expert laparoscopic surgeons.

Wireless live streaming video of laparoscopic surgery: a bandwidth analysis for handheld computers.

Over the last six years, streaming media has emerged as a powerful tool for delivering multimedia content over networks. Concurrently, wireless technology has evolved, freeing users from desktop boundaries and wired infrastructures. At the University of Kentucky Medical Center, we have integrated these technologies to develop a system that can wirelessly transmit live surgery from the operating room to a handheld computer. This study establishes the feasibility of using our system to view surgeries and describes the effect of bandwidth on image quality. A live laparoscopic ventral hernia repair was transmitted to a single handheld computer using five encoding speeds at a constant frame rate, and the quality of the resulting streaming images was evaluated. No video images were rendered when video data were encoded at 28.8 kilobytes per second (Kbps), the slowest encoding bitrate studied. The highest quality images were rendered at encoding speeds greater than or equal to 150 Kbps. Of note, a 15 second transmission delay was experienced using all four encoding schemes that rendered video images. We believe that the wireless transmission of streaming video to handheld computers has tremendous potential to enhance surgical education. For medical students and residents, the ability to view live surgeries, lectures, courses and seminars on handheld computers means a larger number of learning opportunities. In addition, we envision that wireless enabled devices may be used to telemonitor surgical procedures. However, bandwidth availability and streaming delay are major issues that must be addressed before wireless telementoring becomes a reality.

Advancements in immersive VR as a tool for preoperative planning for laparoscopic surgery.

The utility of three-dimensional (3D) models for planning laparoscopic surgery and surgical training has been demonstrated. (1) Computed tomography (CT) scans with oral and intravenous contrast medium are frequently used for preoperative evaluation of patients undergoing complex laparoscopic surgery. Immersive 3D VR overcomes many of the conceptual limitations encountered when conveying or teaching 3D relationships via 2D images traditionally produced by these scans. Over the past year we have made advancements in several areas. First, we have improved the quality of our datasets by utilizing higher resolution multi-detector scans and altering the protocols used. Second, we now register multiple isosurface views with standard axial views and volume textured views to provide additional information and perspective. Third, we now routinely use auto-segmentation techniques to visualize individual structures.