Machine learning model for predicting out-of-hospital cardiac arrests using meteorological and chronological data

Heart - Tập 107 Số 13 - Trang 1084-1091 - 2021
Takahiro Nakashima1,2, Soshiro Ogata2, Teruo Noguchi3, Yoshio Tahara4, Daisuke Onozuka2, Satoshi Kato5, Yoshiki Yamagata6, Sunao Kojima7, Taku Iwami8, Tetsuya Sakamoto9, Ken Nagao10, Hiroshi Nonogi11, Satoshi Yasuda12, Koji Iihara13, Robert W. Neumar1, Kunihiro Nishimura2
1Department of Emergency Medicine, University of Michigan, Ann Arbor, Michigan, USA
2Department of Preventive Medicine and Epidemiologic Informatics, National Cerebral Cardiovascular Centre, Suita, Japan
3NCVC, Suita, Osaka 564-8565, Japan
4Department of Cardiovascular Medicine, National Cerebral and Cardiovascular Centre, Suita, Japan
5H.U. Group Research Institute G.K, Tokyo, Japan
6National Institute for Environmental Studies, Tsukuba, Japan
7Department of General Internal Medicine 3, Kawasaki Medical School, Kurashiki, Japan
8Health Service, Kyoto University, Kyoto, Japan
9Department of Emergency Medicine, Teikyo University, Itabashi-ku, Japan
10Cardiovascular Centre, Nihon University Hospital, Tokyo, Japan
11Shizuoka General Hospital, Shizuoka, Japan
12Department of Cardiovascular Medicine, Tohoku University Graduate School of Medicine, Sendai, Japan
13National Cerebral and Cardiovascular Center, Suita, Osaka, Japan

Tóm tắt

ObjectivesTo evaluate a predictive model for robust estimation of daily out-of-hospital cardiac arrest (OHCA) incidence using a suite of machine learning (ML) approaches and high-resolution meteorological and chronological data.MethodsIn this population-based study, we combined an OHCA nationwide registry and high-resolution meteorological and chronological datasets from Japan. We developed a model to predict daily OHCA incidence with a training dataset for 2005–2013 using the eXtreme Gradient Boosting algorithm. A dataset for 2014–2015 was used to test the predictive model. The main outcome was the accuracy of the predictive model for the number of daily OHCA events, based on mean absolute error (MAE) and mean absolute percentage error (MAPE). In general, a model with MAPE less than 10% is considered highly accurate.ResultsAmong the 1 299 784 OHCA cases, 661 052 OHCA cases of cardiac origin (525 374 cases in the training dataset on which fourfold cross-validation was performed and 135 678 cases in the testing dataset) were included in the analysis. Compared with the ML models using meteorological or chronological variables alone, the ML model with combined meteorological and chronological variables had the highest predictive accuracy in the training (MAE 1.314 and MAPE 7.007%) and testing datasets (MAE 1.547 and MAPE 7.788%). Sunday, Monday, holiday, winter, low ambient temperature and large interday or intraday temperature difference were more strongly associated with OHCA incidence than other the meteorological and chronological variables.ConclusionsA ML predictive model using comprehensive daily meteorological and chronological data allows for highly precise estimates of OHCA incidence.

Từ khóa


Tài liệu tham khảo

10.1161/CIR.0000000000000757

10.1016/j.resuscitation.2010.08.006

10.1053/euhj.1999.1739

10.1161/01.CIR.100.15.1630

10.1016/S0300-9572(02)00097-7

10.1016/j.ajem.2010.06.018

10.1016/j.amjcard.2016.11.041

10.1161/01.CIR.100.1.e1

10.1161/CIRCULATIONAHA.108.815860

10.1016/S0140-6736(96)12338-2

10.1088/1361-6579/aadf0f

10.1093/eurheartj/ehz565

10.1007/s40708-017-0065-7

10.1056/NEJMoa0906644

Chen TQ , Guestrin C . XGBoost: a scalable tree boosting system. Kdd'16: Proceedings of the 22nd Acm Sigkdd International Conference on Knowledge Discovery and Data Mining, 2016:785–94.

Natekin, 2013, Gradient boosting machines, a tutorial, Front Neurorobot, 7, 10.3389/fnbot.2013.00021

10.1161/01.CIR.0000147236.85306.15

10.1056/NEJMsa1600011

10.1161/CIRCULATIONAHA.115.018788

10.1253/circj.CJ-13-0457

10.1016/S0140-6736(19)32488-2

10.1371/journal.pone.0138765

Lundberg SM , Lee SI . A unified approach to interpreting model predictions. Adv Neur In 2017;30.

10.1136/hrt.69.5.385

10.1111/j.1440-1681.1984.tb00254.x

10.1016/0002-9343(86)90348-7

10.1016/S0140-6736(20)30813-8

10.1093/eurheartj/ehy790

Thông tin
Thông tin xuất bản

Heart

Tập 107 Số 13

1084-1091

Thông tin tác giả