Machine-learning-based prognostic models for independence in toilet-related activities in patients with subacute stroke: a retrospective study
SVM provided the highest accuracy with minimal overlearning. LR offered similar performance and greater interpretability, supporting its clinical use. These models could provide valuable information in stroke rehabilitation.
Authors
- Yuta Miyazaki (ORCID: https://orcid.org/0000-0003-2180-5978)
- Tomonori Akimoto
- Michiyuki Kawakami (ORCID: https://orcid.org/0000-0001-6459-3325)
- Tetsuya Tsuji (ORCID: https://orcid.org/0000-0002-4728-9409)
- Takayuki Kamimoto (ORCID: https://orcid.org/0000-0002-2135-1809)
- Nanako Hijikata (ORCID: https://orcid.org/0000-0002-7911-8567)
- Kunitsugu Kondo (ORCID: https://orcid.org/0000-0002-6614-8142)
- Kaoru Honaga (ORCID: https://orcid.org/0000-0003-0885-2876)
- Akiko Hirabe
- Kanjiro Suzuki
- Masahiro Tsujikawa
Institutions
- Keio University (JP)
- Alameda Hospital (US)
- Juntendo University (JP)
- Tokyo Medical Center (JP)
- National Center of Neurology and Psychiatry (JP)
- Hebei Rehabilitation Hospital (JP)
- National Cancer Center Hospital East (JP)
Publication Details
- Journal
- Topics in Stroke Rehabilitation
- Published
- 2025-06-11
- DOI
- https://doi.org/10.1080/10749357.2025.2516850
- Primary Topic
- Acute Ischemic Stroke Management
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- France Bed Medical Home Care Research Subsidy Foundation
- Japan Society for the Promotion of Science