Fully-automatic software to estimate optimal resected regions considering practical conditions for liver surgery

Abstract

Aim: Recent advances in three-dimensional (3D) imaging have been remarkable, and 3D simulation has become an indispensable item in liver surgery. Furthermore, image processing is currently shifting from simulation software (SM) to NAVIGATIONS SOFTWARE (NS) development. The Aim of this study is to report on a new version of SM (Hariyama Shimoda software; HSS) with added preoperative liver function evaluation, and to report the results of NS development using 3D ultrasound (US) images. Results: We evaluated 21 cases with solitary liver tumors that underwent hepatic resection between February 2021 and June 2023 in our department. Estimated liver resection volume was positively correlated between Synaptic Vincent (SV) conventional SM software and newly developed HSS software (p < 0.001 r: 0.813), and actual resected liver volume was also positively correlated between SV and HSS (p = 0.002 r: 0.467, p = 0.002 r: 0. 641, respectively). In addition, by using the magnetic sensor “AURORA”, we were able to produce 3DUS images and were able to extract PV from US images using deep learning and region growing. Conclusion: HSS is as effective a 3D NS as SV, and 3D US could be used as navigation for hepatectomy in the future.

Keywords

• Three-dimensional CT simulation
• Three-dimensional CT simulation full automations
• Navigation