Development of reliable daily physical function evaluation system using model-based simple measurement method

2022 Fiscal Year Final Research Report

• Project/Area Number: 17K00230
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Perceptual information processing
• Research Institution: Yokohama National University
• Principal Investigator: Sugimoto Chika 横浜国立大学, 大学院工学研究院, 准教授 (40447347)
• Project Period (FY): 2017-04-01 – 2023-03-31
• Keywords: 咀嚼・嚥下 / 呼吸 / 機能評価 / 深度情報 / RGB画像 / 機械学習

Outline of Final Research Achievements

We proposed a new reliable method for evaluating the important human functions of chewing, swallowing, and breathing using an RGB-D camera. The evaluation models were developed based on the 3D feature points of the face, neck, and body obtained from RGB and depth information. The path, speed, and rhythm of the movement of occlusion and mouth opening and closing and the volume change of the occlusal muscles for each cycle are evaluated for mastication. The movement trajectory of pharyngeal prominence is tracked by extracting 3D features of the laryngeal region and detecting the prominence for evaluation of swallowing. The degree of thoracic and abdominal breathing is calculated based on the changes in volume of the chest and abdomen by detecting the movement of the diaphragm from the costal arch for evaluation of breathing. As the result, the methods showed that the states and functions of chewing, swallowing, and breathing can be evaluated with high accuracy using a simple device.

Free Research Field

人間情報学、ヒューマンセンシング

Academic Significance and Societal Importance of the Research Achievements

高価な専用装置を用いず、簡易に導入可能な機器の測定データを基に、先端的かつ高度な機械学習手法やデータ解析手法、有用な生体モデルを活用することで、口や顎周辺、頸部の喉頭隆起、胸腹部の細かな動きを高精度に検出し、非侵襲・非接触に咀嚼・嚥下、呼吸機能を客観的に評価する手法は新規性があり学術的意義がある。
また、簡易計測手法の構築により経時的データ取得が容易になるため、機能低下の特徴や原因の分析、予防・訓練手法の開発に有用であり、健康維持、疾患予防や早期発見にもつながることから社会的意義も高い。