超音波動画内の正中神経セグメンテーションと手根管症候群推定

• Item type: SIG Technical Reports(1)
• 公開日: 2022-09-29
• タイトル: 超音波動画内の正中神経セグメンテーションと手根管症候群推定
• タイトル / 言語: en
• タイトル: Median Nerve Segmentation and Carpal Tunnel Syndrome Estimation in Ultrasound Movies
• 言語: jpn
• キーワード / 主題Scheme: Other
• キーワード / 主題: 計測・認証1
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: 慶應義塾大学
• 著者所属: 慶應義塾大学
• 著者所属: 東京医科歯科大学
• 著者所属: 東京医科歯科大学
• 著者所属: 東京医科歯科大学
• 著者所属: 慶應義塾大学
• 著者所属(英): en
• 著者所属(英): Keio University
• 著者所属(英): en
• 著者所属(英): Keio University
• 著者所属(英): en
• 著者所属(英): Tokyo Medical And Dental University
• 著者所属(英): en
• 著者所属(英): Tokyo Medical And Dental University
• 著者所属(英): en
• 著者所属(英): Tokyo Medical And Dental University
• 著者所属(英): en
• 著者所属(英): Keio University
• 著者名: 佐藤, 優希菜; 松尾, 佳奈; 小山, 恭史; 山田, 英莉久; 藤田, 浩二; 杉浦, 裕太
• 著者名(英): Yukina, Sato; Kana, Matsuo; Takafumi, Koyama; Eriku, Yamada; Koji, Fujita; Yuta, Sugiura
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: Carpal tunnel syndrome is a disease that causes numbness and motor dysfunction of the fingers due to compression of the median nerve. In this study, Mask R-CNN is used to estimate the area of the nerve in each frame of an ultrasound movie of a specific hand movement. Features of the estimated regions are extracted by image processing, and time series data such as convex area, perimeter, flattening, center-of-gravity coordinates and eccentricity are collected to estimate the presence and severity of carpal tunnel syndrome. Class classification based on group k-fold cross-validation using features obtained from 37 videos of patients and 22 videos of healthy subjects resulted in an accuracy rate of 66.1%, sensitivity of 82.9%, and specificity of 79.2%.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: Carpal tunnel syndrome is a disease that causes numbness and motor dysfunction of the fingers due to compression of the median nerve. In this study, Mask R-CNN is used to estimate the area of the nerve in each frame of an ultrasound movie of a specific hand movement. Features of the estimated regions are extracted by image processing, and time series data such as convex area, perimeter, flattening, center-of-gravity coordinates and eccentricity are collected to estimate the presence and severity of carpal tunnel syndrome. Class classification based on group k-fold cross-validation using features obtained from 37 videos of patients and 22 videos of healthy subjects resulted in an accuracy rate of 66.1%, sensitivity of 82.9%, and specificity of 79.2%.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AA12049625
• 書誌情報: 研究報告エンタテインメントコンピューティング（EC）; 巻 2022-EC-65, 号 18, p. 1-4, 発行日 2022-09-29
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 2188-8914
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会