Realization of Highly Visible Surgical Assistance Imagery through the Integration of Surgical Microscope and OCT Information

2022 Fiscal Year Final Research Report

• Project/Area Number: 21K18074
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 90130:Medical systems-related
• Research Institution: The University of Tokyo
• Principal Investigator: Sogabe Maina 東京大学, 大学院情報理工学系研究科, 助教 (80788951)
• Project Period (FY): 2021-04-01 – 2023-03-31
• Keywords: 眼科 / OCT / VR

Outline of Final Research Achievements

This research aimed to innovate ophthalmic surgery techniques through the aggregation of various intraoperative information. During this aggregation, it was found that various intraoperative noises posed obstacles, hence frameworks utilizing deep learning were developed to address each problem. The research included the development of an environment for the acquisition and presentation of three-dimensional information using camera and OCT data. This allowed for the real-time reconstruction of depth information and facilitated the rendering and annotation within VR goggles, establishing the foundation for intraoperative information aggregation. Additionally, technologies for recognizing the location of bleeding and ophthalmic instruments were developed using deep learning to tackle the issue of visual field obstruction. The combination of these techniques is believed to contribute to the improvement of accuracy and efficiency in ophthalmic surgery.

Free Research Field

医療画像処理

Academic Significance and Societal Importance of the Research Achievements

この研究の学術的意義は、医療技術とAIの融合を具現化する新たな枠組みを提供し、眼科手術の技術革新を牽引する重要な一歩となっている。OCTを用いたロボット制御、深層学習の活用、3次元情報の取得と提示、ノイズ除去技術の開発は、未来の手術環境の可能性を広げている。眼科手術の精度と効率性の向上は、患者の治療結果改善と医療資源の最適化を可能にする。手術の安全性を向上し、医師の労働負荷を軽減するロボット技術とAIの組み合わせは、医療現場の進化につながることが期待されており、高齢化社会における眼科疾患の増加に対応し、社会全体の健康と福祉の向上に貢献することが期待される。