{"id":216640,"updated":"2025-01-19T15:47:08.244237+00:00","links":{},"created":"2025-01-19T01:17:09.773443+00:00","metadata":{"_oai":{"id":"oai:ipsj.ixsq.nii.ac.jp:00216640","sets":["1164:5159:10869:10870"]},"path":["10870"],"owner":"44499","recid":"216640","title":["リーマン多様体による能動学習を用いたてんかん発作検出"],"pubdate":{"attribute_name":"公開日","attribute_value":"2022-02-22"},"_buckets":{"deposit":"568256df-8fa6-4d6e-9aed-09befaec846e"},"_deposit":{"id":"216640","pid":{"type":"depid","value":"216640","revision_id":0},"owners":[44499],"status":"published","created_by":44499},"item_title":"リーマン多様体による能動学習を用いたてんかん発作検出","author_link":["559463","559464","559465","559466"],"item_titles":{"attribute_name":"タイトル","attribute_value_mlt":[{"subitem_title":"リーマン多様体による能動学習を用いたてんかん発作検出"},{"subitem_title":"Epileptic Seizure Detection Using Active Learning with Riemannian Manifold","subitem_title_language":"en"}]},"item_keyword":{"attribute_name":"キーワード","attribute_value_mlt":[{"subitem_subject":"ポスターセッション3","subitem_subject_scheme":"Other"}]},"item_type_id":"4","publish_date":"2022-02-22","item_4_text_3":{"attribute_name":"著者所属","attribute_value_mlt":[{"subitem_text_value":"東京農工大学大学院工学府電気電子工学専攻"},{"subitem_text_value":"東京農工大学大学院工学府電気電子工学専攻"}]},"item_4_text_4":{"attribute_name":"著者所属(英)","attribute_value_mlt":[{"subitem_text_value":"Toshiki Orihara","subitem_text_language":"en"},{"subitem_text_value":"Toshihisa Tanaka","subitem_text_language":"en"}]},"item_language":{"attribute_name":"言語","attribute_value_mlt":[{"subitem_language":"jpn"}]},"item_publisher":{"attribute_name":"出版者","attribute_value_mlt":[{"subitem_publisher":"情報処理学会","subitem_publisher_language":"ja"}]},"publish_status":"0","weko_shared_id":-1,"item_file_price":{"attribute_name":"Billing file","attribute_type":"file","attribute_value_mlt":[{"url":{"url":"https://ipsj.ixsq.nii.ac.jp/record/216640/files/IPSJ-SLP22140039.pdf","label":"IPSJ-SLP22140039.pdf"},"format":"application/pdf","billing":["billing_file"],"filename":"IPSJ-SLP22140039.pdf","filesize":[{"value":"1.7 MB"}],"mimetype":"application/pdf","priceinfo":[{"tax":["include_tax"],"price":"0","billingrole":"22"},{"tax":["include_tax"],"price":"0","billingrole":"44"}],"accessrole":"open_login","version_id":"0bf79324-9518-4ae0-8405-df8eb063cebd","displaytype":"detail","licensetype":"license_note","license_note":"Copyright (c) 2022 by the Institute of Electronics, Information and Communication Engineers This SIG report is only available to those in membership of the SIG."}]},"item_4_creator_5":{"attribute_name":"著者名","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"折原, 俊貴"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"田中, 聡久"}],"nameIdentifiers":[{}]}]},"item_4_creator_6":{"attribute_name":"著者名(英)","attribute_type":"creator","attribute_value_mlt":[{"creatorNames":[{"creatorName":"Toshiki, Orihara","creatorNameLang":"en"}],"nameIdentifiers":[{}]},{"creatorNames":[{"creatorName":"Toshihisa, Tanaka","creatorNameLang":"en"}],"nameIdentifiers":[{}]}]},"item_4_source_id_9":{"attribute_name":"書誌レコードID","attribute_value_mlt":[{"subitem_source_identifier":"AN10442647","subitem_source_identifier_type":"NCID"}]},"item_4_textarea_12":{"attribute_name":"Notice","attribute_value_mlt":[{"subitem_textarea_value":"SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc."}]},"item_resource_type":{"attribute_name":"資源タイプ","attribute_value_mlt":[{"resourceuri":"http://purl.org/coar/resource_type/c_18gh","resourcetype":"technical report"}]},"item_4_source_id_11":{"attribute_name":"ISSN","attribute_value_mlt":[{"subitem_source_identifier":"2188-8663","subitem_source_identifier_type":"ISSN"}]},"item_4_description_7":{"attribute_name":"論文抄録","attribute_value_mlt":[{"subitem_description":"機械学習による診断支援技術を実現するには，患者ごとにデータの分布が異なるため，訓練モデルが必ずしも未知のデータに対して有効ではない問題（ドメイン適応問題）を解決しなくてはならない．本稿では，リーマン多様体上の平行移動による教師なしドメイン適応の手法を，てんかん患者の脳波に対する発作検出に適用する．その際，ターゲットドメインからリーマン距離の近いソースドメインの中心点に向けて，ソースデータとターゲットデータの双方を平行移動させることで，モデルを能動的に訓練する仕組みを提案する．発作時脳波を含む公開データセットを用いて患者間検証を実施したところ，提案手法を用いることで高い AUC 及び正解率で発作時脳波を検出できることを確認した．本稿で提案する手法は，脳波に対する発作検出だけでなく，ドメインシフトが想定される脳波の識別全般に応用できることが示唆される．","subitem_description_type":"Other"}]},"item_4_description_8":{"attribute_name":"論文抄録(英)","attribute_value_mlt":[{"subitem_description":"In order to realize machine learning for diagnosis, it is necessary to solve the problem that the training model is not always effective for unknown data (domain adaptation problem) because the distribution of data is different for each patient. In this paper, we apply the unsupervised domain adaptation method by translation on a Riemannian manifold to seizure detection for EEG of epilepsy patients. We propose a mechanism to actively train the model by translating both the source and target data to the centre point of the source domain, which is close to the Riemannian distance from the target domain. We conducted inter-patient validation using a public dataset including seizure EEG, and confirmed that the proposed method can detect seizure EEG with high AUC and correct answer rate. It is suggested that the proposed method can be applied not only to seizure detection for EEG but also to the identification of EEG with domain shift in general.","subitem_description_type":"Other"}]},"item_4_biblio_info_10":{"attribute_name":"書誌情報","attribute_value_mlt":[{"bibliographicPageEnd":"6","bibliographic_titles":[{"bibliographic_title":"研究報告音声言語情報処理（SLP）"}],"bibliographicPageStart":"1","bibliographicIssueDates":{"bibliographicIssueDate":"2022-02-22","bibliographicIssueDateType":"Issued"},"bibliographicIssueNumber":"39","bibliographicVolumeNumber":"2022-SLP-140"}]},"relation_version_is_last":true,"weko_creator_id":"44499"}}