Feasibility of Simultaneous Artificial Intelligence-Assisted and NIR Fluorescence Navigation for Anatomical Recognition in Laparoscopic Colorectal Surgery.

Abstract

Ureters can be visualized on a monitor via fluorescence observation technology and an near-infrared (NIR) fluorescent ureteral catheter (NIRFUC). Eureka, an artificial intelligence (AI) platform, can be used to analyze surgical videos and highlight nerves and loose connective tissue (LCT) in the dissection layer. In this study, we aimed to evaluate the feasibility of using simultaneous NIRFUC and AI assistance for anatomical recognition during laparoscopic surgery. The research target was video recordings of laparoscopic colorectal surgery in which the ureters were visualized using an NIRFUC (n = 56, November 2022 to May 2024). Eureka was used to analyze the nerves and LCTs in these videos. Three physicians reviewed and analyzed the videos, scoring the fluorescence visualization of the ureters and LCT by Eureka, the fluorescence visualization of the ureters and hypogastric nerve by Eureka, and the fluorescence visualization of the ureters and lumbar splanchnic nerves by Eureka, using a Likert scale. The scoring system was as follows: 0, very poor; 1, poor; 2, acceptable; 3, good; and 4, very good. The mean Likert scale score was 3.99 for the ureters and LCT, 3.11 for the ureters and hypogastric nerve, and 3.53 for the ureters and lumbar splanchnic nerves. The training data used for this AI model did not include NIR fluorescence image observations. Anatomical highlighting with AI and fluorescence visualization of the ureters were possible in the images analyzed by Eureka. These findings suggest that both AI and NIR can be used simultaneously for real-time navigation in the future.