Machine learning-based prediction models for accidental hypothermia patients

Okada, Yohei; Matsuyama, Tasuku; Morita, Sachiko; Ehara, Naoki; Miyamae, Nobuhiro; Jo, Takaaki; Sumida, Yasuyuki; Okada, Nobunaga; Watanabe, Makoto; Nozawa, Masahiro; Tsuruoka, Ayumu; Fujimoto, Yoshihiro; Okumura, Yoshiki; Kitamura, Tetsuhisa; Iiduka, Ryoji; Ohtsuru, Shigeru

doi:https://doi.org/10.1186/s40560-021-00525-z

Machine learning-based prediction models for accidental hypothermia patients

Okada, Yohei; Matsuyama, Tasuku; Morita, Sachiko et al.

Journal of Intensive Care, 2021, 9, 6

アクセス数:1,437件（2025-12-10 11:01 集計）

固定URL: https://hdl.handle.net/11094/78364

閲覧可能ファイル

ファイル	フォーマット	利用条件	サイズ	閲覧回数	利用開始日	説明	information
JIntensiveCare_09_006	pdf	なし	1.71 MB	195	2021.02.02

Creative Commons : 表示

論文情報

ファイル出力 EndNote Basic出力 Mendeley出力

タイトル	Machine learning-based prediction models for accidental hypothermia patients
著者	著者名 Okada, Yohei
	NRID 1000050829025 1000050829025 著者名 Matsuyama, Tasuku
	著者名 Morita, Sachiko
	著者名 Ehara, Naoki
	著者名 Miyamae, Nobuhiro
	著者名 Jo, Takaaki
	著者名 Sumida, Yasuyuki
	著者名 Okada, Nobunaga
	著者名 Watanabe, Makoto
	著者名 Nozawa, Masahiro
	著者名 Tsuruoka, Ayumu
	著者名 Fujimoto, Yoshihiro
	著者名 Okumura, Yoshiki
	NRID 1000030639810 1000030639810 ORCID 0000-0003-0107-0580 0000-0003-0107-0580 researchmap permalink Kitamura_T Kitamura_T Scopus Author ID 22234282700 22234282700 著者名 Kitamura, Tetsuhisa
	著者名 Iiduka, Ryoji
	NRID 1000060437225 1000060437225 著者名 Ohtsuru, Shigeru
キーワード等	Accidental hypothermia
	Artificial intelligence
	Gradient boosting tree
	Lasso
	Machine learning
	Prediction
	Random forest
内容	Okada, Y., Matsuyama, T., Morita, S., et al. Machine learning-based prediction models for accidental hypothermia patients. j intensive care 9, 6 (2021). https://doi.org/10.1186/s40560-021-00525-z
抄録	Background: Accidental hypothermia is a critical condition with high risks of fatal arrhythmia, multiple organ failure, and mortality; however, there is no established model to predict the mortality. The present study aimed to develop and validate machine learning-based models for predicting in-hospital mortality using easily available data at hospital admission among the patients with accidental hypothermia. Method: This study was secondary analysis of multi-center retrospective cohort study (J-point registry) including patients with accidental hypothermia. Adult patients with body temperature 35.0 °C or less at emergency department were included. Prediction models for in-hospital mortality using machine learning (lasso, random forest, and gradient boosting tree) were made in development cohort from six hospitals, and the predictive performance were assessed in validation cohort from other six hospitals. As a reference, we compared the SOFA score and 5A score. Results: We included total 532 patients in the development cohort [N = 288, six hospitals, in-hospital mortality: 22.0% (64/288)], and the validation cohort [N = 244, six hospitals, in-hospital mortality 27.0% (66/244)]. The C-statistics [95% CI] of the models in validation cohorts were as follows: lasso 0.784 [0.717-0.851] , random forest 0.794[0.735-0.853], gradient boosting tree 0.780 [0.714-0.847], SOFA 0.787 [0.722-0.851], and 5A score 0.750[0.681-0.820]. The calibration plot showed that these models were well calibrated to observed in-hospital mortality. Decision curve analysis indicated that these models obtained clinical net-benefit. Conclusion: This multi-center retrospective cohort study indicated that machine learning-based prediction models could accurately predict in-hospital mortality in validation cohort among the accidental hypothermia patients. These models might be able to support physicians and patient’s decision-making. However, the applicability to clinical settings, and the actual clinical utility is still unclear; thus, further prospective study is warranted to evaluate the clinical usefulness.
出版者	BioMed Central Ltd
収録物名	Journal of Intensive Care
巻(号)	9
ページ	6
刊行年月	2021-01-09
言語	英語
ハンドルURL	https://hdl.handle.net/11094/78364
eISSN	20520492
関連情報 (isIdenticalTo)	DOI (出版社版) https://doi.org/10.1186/s40560-021-00525-z
アクセス権	open access
権利情報	© 2021 The Author(s). This article is licensed under a Creative Commons Attribution 4.0 International License.
出版タイプ	Version of Record
カテゴリ	学術雑誌論文 Journal Article

資源タイプ	学術雑誌論文
ローカル資源タイプ	学術雑誌論文
DCMI資源タイプ	text
DCTERMS.bibliographicCitation	Journal of Intensive Care. 2021, 9, p. 6
DC.title	Machine learning-based prediction models for accidental hypothermia patients
DC.creator	Okada, Yohei
	Matsuyama, Tasuku
	Morita, Sachiko
	Ehara, Naoki
	Miyamae, Nobuhiro
	Jo, Takaaki
	Sumida, Yasuyuki
	Okada, Nobunaga
	Watanabe, Makoto
	Nozawa, Masahiro
	Tsuruoka, Ayumu
	Fujimoto, Yoshihiro
	Okumura, Yoshiki
	Kitamura, Tetsuhisa
	Iiduka, Ryoji
	Ohtsuru, Shigeru
DC.publisher	BioMed Central Ltd
DC.language' scheme='DCTERMS.RFC1766	英語
DC.type' scheme='DCTERMS.DCMIType	text
DCTERMS.issued' scheme='DCTERMS.W3CDTF	2021-01-09
DC.identifier	https://doi.org/10.1186/s40560-021-00525-z
DC.subject	Accidental hypothermia
	Artificial intelligence
	Gradient boosting tree
	Lasso
	Machine learning
	Prediction
	Random forest
DC.description	Okada, Y., Matsuyama, T., Morita, S., et al. Machine learning-based prediction models for accidental hypothermia patients. j intensive care 9, 6 (2021). https://doi.org/10.1186/s40560-021-00525-z
DCTERMS.abstract	Background: Accidental hypothermia is a critical condition with high risks of fatal arrhythmia, multiple organ failure, and mortality; however, there is no established model to predict the mortality. The present study aimed to develop and validate machine learning-based models for predicting in-hospital mortality using easily available data at hospital admission among the patients with accidental hypothermia. Method: This study was secondary analysis of multi-center retrospective cohort study (J-point registry) including patients with accidental hypothermia. Adult patients with body temperature 35.0 °C or less at emergency department were included. Prediction models for in-hospital mortality using machine learning (lasso, random forest, and gradient boosting tree) were made in development cohort from six hospitals, and the predictive performance were assessed in validation cohort from other six hospitals. As a reference, we compared the SOFA score and 5A score. Results: We included total 532 patients in the development cohort [N = 288, six hospitals, in-hospital mortality: 22.0% (64/288)], and the validation cohort [N = 244, six hospitals, in-hospital mortality 27.0% (66/244)]. The C-statistics [95% CI] of the models in validation cohorts were as follows: lasso 0.784 [0.717-0.851] , random forest 0.794[0.735-0.853], gradient boosting tree 0.780 [0.714-0.847], SOFA 0.787 [0.722-0.851], and 5A score 0.750[0.681-0.820]. The calibration plot showed that these models were well calibrated to observed in-hospital mortality. Decision curve analysis indicated that these models obtained clinical net-benefit. Conclusion: This multi-center retrospective cohort study indicated that machine learning-based prediction models could accurately predict in-hospital mortality in validation cohort among the accidental hypothermia patients. These models might be able to support physicians and patient’s decision-making. However, the applicability to clinical settings, and the actual clinical utility is still unclear; thus, further prospective study is warranted to evaluate the clinical usefulness.
DC.rights	© 2021 The Author(s). This article is licensed under a Creative Commons Attribution 4.0 International License.
DC.format	78364
citation_title	Machine learning-based prediction models for accidental hypothermia patients
citation_author	Okada, Yohei
	Matsuyama, Tasuku
	Morita, Sachiko
	Ehara, Naoki
	Miyamae, Nobuhiro
	Jo, Takaaki
	Sumida, Yasuyuki
	Okada, Nobunaga
	Watanabe, Makoto
	Nozawa, Masahiro
	Tsuruoka, Ayumu
	Fujimoto, Yoshihiro
	Okumura, Yoshiki
	Kitamura, Tetsuhisa
	Iiduka, Ryoji
	Ohtsuru, Shigeru
citation_publisher	BioMed Central Ltd
citation_language	英語
citation_date	2021-01-09
citation_journal_title	Journal of Intensive Care
citation_volume	9
citation_firstpage	6
citation_doi	https://doi.org/10.1186/s40560-021-00525-z
citation_public_url	https://hdl.handle.net/11094/78364

Machine learning-based prediction models for accidental hypothermia patients

閲覧可能ファイル

論文情報

論文詳細を表示