Development of video analysis based artificial intelligence for hemifacial spasm care

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K16623
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 56010:Neurosurgery-related
• Research Institution: Hokkaido University
• Principal Investigator: ITO YASUHIRO 北海道大学, 医学研究院, 客員研究員 (80899310)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: 片側顔面痙攣 / 動画解析

Outline of Final Research Achievements

I In this study, we conducted simultaneous electromyography and video recordings of facial expression muscles in patients with involuntary facial movements. Using video analysis software, we carried out fixed-point tracking at three locations corresponding to the orbicularis oculi, orbicularis oris, and zygomaticus muscles. Our initial analysis of 17 patients with hemifacial spasms revealed that involuntary facial movements, observed during the examination, could be depicted as waveforms with short peaks or plateaus confined to the affected side, based on displacement analysis. Furthermore, we were able to numerically quantify this displacement on both the X and Y axes, allowing us to quantify the movements of facial muscles in patients with hemifacial spasms.
In conclusion, a simple and minimally invasive video analysis method can be applicable for distinguishing between other involuntary facial movement disorders and assessing the effectiveness of various treatments.

Free Research Field

脳神経外科

Academic Significance and Societal Importance of the Research Achievements

本研究成果から、簡便で低侵襲なビデオ分析手法により、片側顔面痙攣における特徴量を定量分析し得ることが示唆された。他の不随意顔面運動疾患（心因性顔面運動、顔面ミオキミア、眼瞼痙攣、チック、遅発性ジスキネジア）との鑑別診断補助、A型ボツリヌス毒素注射部位決定の補助や、各種治療効果の判定にも応用が可能であるという医学的な意義がある。