Modeling and removing visual impairments for endoscopic kidney stone surgery project summary/abstract the importance of achieving stone-free status during endoscopic kidney stone surgery is emphasized…

The National Institute of Diabetes and Digestive and Kidney Diseases awarded Vanderbilt University $149,190 on September 15, 2025, under the Diabetes, Digestive, and Kidney Diseases Extramural Research program (CFDA 93.847) to develop computer vision models that quantify and remove visual impairments during endoscopic kidney stone surgery.

The project addresses the high rate of repeat stone procedures caused by residual stone fragments after initial surgery. Blood and debris frequently obscure the surgeon's field of view during endoscopic surgery, impairing stone visualization and location. The funded research will pursue two specific objectives: an automatic, real-time method for quantifying visual impairments and associating them with surgical outcomes; and digital models for virtual removal of visual impairments during surgery. The deliverables include an objective quantification method for visual impairments and real-time digital removal capability from endoscopic stone surgery video. Success could decrease residual fragment frequency, reduce repeat surgeries, and improve treatment efficiency by reducing operative time and injury risk to the collecting system.

Performance occurs at Vanderbilt University in Nashville, Tennessee, with a project period through September 14, 2028. The award is classified as a Project Grant and represents assistance funding.

Modeling and removing visual impairments for endoscopic kidney stone surgery project summary/abstract 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. Residual stone fragments are caused by incomplete stone treatment and can lead to obstruction, pain, kidney injury, and recurrent infections. Among other issues, blood and debris can frequently obscure the already limited field of view during endoscopic stone surgery, impairing the surgeon's ability to visualize kidney stones. Importantly, endoscopic video is the primary means of locating and tracking stones during treatment. Though technical constraints of scopes may also impact surgical kidney stone outcomes, the above limitations of endoscopic stone treatment prevent many surgeons from achieving a complete stone-free status. My goal is to test whether computer vision models can be used to quantify and remove visual impairments during kidney stone surgery. I propose two specific aims to achieve these two objectives: aim 1 will develop an automatic, real-time method for quantifying visual impairments and seek to associate impairments with surgical outcomes. Aim 2 will develop models for the virtual removal of visual impairments during surgery. These studies will produce a novel method for the objective quantification of visual impairments during surgery and means for the real-time digital removal of visual impairments from endoscopic stone surgery video. Success in visual impairment removal could decrease the frequency of residual fragments and reduce the risk of repeat stone surgeries. IT could also lead to more efficient stone treatment, reducing operative time and decreasing the risk of injury to the collecting system. As these models would require only software integration to deploy in real-time on current endoscopic surgical cameras, all existing endoscopic surgical systems could in principle immediately benefit from the results of this project.