Predictive Performance of Machine Learning–Based Models for Poststroke Clinical Outcomes in Comparison With Conventional Prognostic Scores: Multicenter, Hospital-Based Observational Study

Fumi Irie   ^{1, 2 *} , MD, PhD ;   Koutarou Matsumoto   ^{3 *} , MPH, PhD ;   Ryu Matsuo   ^{1, 2} , MD, PhD ;   Yasunobu Nohara   ⁴ , PhD ;   Yoshinobu Wakisaka   ² , MD, PhD ;   Tetsuro Ago   ^{2, 5} , MD, PhD ;   Naoki Nakashima   ⁶ , MD, PhD ;   Takanari Kitazono   ^{2, 5} , MD, PhD ;   Masahiro Kamouchi   ^{1, 5} , MD, PhD

¹ Department of Health Care Administration and Management, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

² Department of Medicine and Clinical Science, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

³ Biostatistics Center, Graduate School of Medicine, Kurume University, Kurume, Japan

⁴ Big Data Science and Technology, Faculty of Advanced Science and Technology, Kumamoto University, Kumamoto, Japan

⁵ Center for Cohort Studies, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan

⁶ Medical Information Center, Kyushu University Hospital, Fukuoka, Japan

*these authors contributed equally