Development of diagnosis support system for mastication and swallowing function applying machine learning by artificial intelligence

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K19091
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 57050:Prosthodontics-related
• Research Institution: Okayama University
• Principal Investigator: Osaka Suguru 岡山大学, 大学病院, 医員 (70823954)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 咀嚼障害 / 人工知能 / 嚥下障害 / 咀嚼運動

Outline of Final Research Achievements

This research developed the application program which evaluate patients’ chewing and swallowing ability from the videos during their food intake. We performed machine learning using CNN and achieved to discriminate the healthy chewing and abnormal chewing. Further, the application program was modified to whose data collection could be performed even by patients’ family or long-term care workers, because face-to-face video data collection was difficult due to the COVID-19 situation.

Free Research Field

高齢者の疫学研究

Academic Significance and Societal Importance of the Research Achievements

認知症や脳血管疾患等で生じる運動障害性咀嚼障害は，一般的な咀嚼・嚥下機能検査での診断が難しく，介護現場で広く使用できるスクリーニングツールもない．今回，我々が開発した咀嚼・嚥下運動評価プログラムを活用すれば，医師・歯科医師が常駐していない老人介護施設や病院，在宅環境でも，運動障害性咀嚼障害や摂食嚥下障害のスクリーニング診断が実装できる可能性がある．さらに，将来的に遠隔医療支援システム等を介してその結果の妥当性を常に専門医がチェックできる体制が構築できれば，国民の健康向上に資する新たな医療システムを創造できると考えられる．