Development and Validation of Gaze-based Training for Endoscopic Kidney Stone Surgery

Project Summary The importance of achieving stone-free status during endoscopic kidney stone surgery is emphasized by the high rate of repeat stone procedures due to residual fragments after index surgery. Specifically, residual stone fragments can lead to obstruction, pain, kidney injury, and recurrent infections. Currently, of the 100,000 patients

who undergo an endoscopic kidney stone treatment each year, around 25,000 will require a repeat stone surgery within 20 months. Successful endoscopic stone surgery requires the surgeon to visualize the entire renal col- lecting system and locate all kidney stones during treatment. Challenges that lead to incomplete stone treatment

involve inadequate stone visibility and difficulty navigating through the kidney. Complete kidney navigation relies on clinical experience and surgeons who have performed fewer cases have greater chances of recurrence. Our overall goal is to create an eye gaze based guidance system to complement the current standard of care

where trainees only get verbal feedback. Gaze guidance has been shown to improve skill acquisition and retention in robot-assisted and laparoscopic surgery settings but has not been explored in the unconstrained environment of ureteroscopy. Our previous work in measuring eye gaze patterns shows similar differences in trainee vs.

expert gaze behavior during kidney stone phantom procedures compared to those found in laparoscopic and robotic surgery. This suggested that using gaze training may similarly improve ureteroscopy skill acquisition. Successful eye tracking and eye gaze sharing in ureteroscopy require a tracking and registration system that

does not constrain surgeons’ movements or head orientations. We propose head-mounted eye trackers that map gaze to the surgical monitor to flexibly track gaze as surgeons move around the operating room. We will evaluate whether the gaze sharing platform improves complete kidney visualization, which is required to identify stones

that may not show up in pre-operative imaging, in phantoms and patients. The endpoint of this R21 will be a fully validated gaze-based training system for endoscopic stone surgery and the necessary experimental data to power a large-scale, multi-center clinical trial. As our training system is solely

composed of software running on augmented reality devices, all existing endoscopic and laparoscopic surgical procedures could in principle immediately benefit from the results of this project. In this way, we believe the suc- cess of our project will facilitate improved training procedures and mitigate repeat interventions or complications,

benefiting patients, surgeons, and society.