Tokyo, June 1 -- UMIN Clinical Trials Registry (UMIN-CTR) received information related to the study (UMIN000061599) titled 'Development of an AI Model to Predict Functional Recovery after Stroke Rehabilitation' on May 31.
Study Type:
Observational
Primary Sponsor:
Institute - Hokkaido University of Science
Condition:
Condition - stroke
Classification by malignancy - Others
Genomic information - NO
Objective:
Narrative objectives1 - To develop a Random Forest model predicting discharge FIM scores in stroke patients in convalescent rehabilitation wards, incorporating aging-related variables, and to evaluate its predictive accuracy and clinical interpretability using SHAP analysis.
Basic objectives2 - Efficacy
Eligibility:
Age-lower limit - Not applicable
Age-upper limit - Not applicable
Gender - Male and Female
Key inclusion criteria - Patients admitted to the convalescent rehabilitation ward at Shin-Sapporo Neurosurgical Hospital between January 2018 and March 2020
Patients with a primary diagnosis of cerebral infarction or intracerebral hemorrhage
Patients with a length of stay in the convalescent rehabilitation ward of at least one month
Key exclusion criteria - Patients with a history of neurological diseases (e.g., Parkinson's disease, multiple sclerosis)
Patients with a history of psychiatric disorders (e.g., schizophrenia, bipolar disorder)
Patients with a history or complication of brain tumor
Patients who experienced clinical deterioration during hospitalization (e.g., stroke recurrence, pneumonia, exacerbation of heart failure)
Patients with blindness
Patients who developed stroke during hospitalization at another medical facility
Patients with missing data
Target Size - 300
Recruitment Status:
Recruitment status - No longer recruiting
Date of protocol fixation - 2022 Year 03 Month 07 Day
Date of IRB - 2022 Year 03 Month 07 Day
Anticipated trial start date - 2022 Year 03 Month 07 Day
Last follow-up date - 2026 Year 04 Month 30 Day
To know more, visit https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000070479
Disclaimer: Curated by HT Syndication.