Classification of Two Type Planarians Behavior Based on Pretrained Deep Learning Models

カテゴリ: 部門大会

論文No: SS3-5

グループ名: 【C】2022年電気学会電子・情報・システム部門大会

発行日: 2022/08/24

タイトル(英語): Classification of Two Type Planarians Behavior Based on Pretrained Deep Learning Models

著者名: 葛城 玲弥（岩手大学）,下條 研輔（ハーバードウェストレイクスクール）,下條 英子（カリフォルニア工科大学）,下條 信輔（カリフォルニア工科大学）,明石 卓也（岩手大学）

著者名(英語): Reiya Katsuragi (Iwate University),Kensuke Shimojo (Harvard-Westlake School),Eiko Shimojo (California Institute of Technology),Shinsuke Shimojo (California Institute of Technology),Takuya Akashi (Iwate University)

キーワード: プラナリア|深層学習|画像分類|イメージセンシング|Planarian|Deep Learning|Image Classification|Image Sensing

要約(日本語): This study classified two types of external-stimulus-induced planarian behavior as the target of observation. Planarians, a type of free-living flatworm, are known for their learning and regenerative abilities, even with the loss of a portion of the head. Thus, they are used as experimental models in organ regeneration studies. Meanwhile, the experimental observation and evaluation of planarian behavior is generally determined by human visual inspection, which makes such evaluation undesirably subjective. Therefore, we propose the video classification of planarian behavior using pre-trained deep learning models for extracting image features from a video frame of video data to provide more objective evidence for biological experiments.