Related Experiment Video
Updated: Jan 9, 2026
07:42
Dual-Task Stroop Paradigm for Detecting Cognitive Deficits in High-Functioning Stroke Patients
Published on: December 16, 2022
3.6K
Using machine learning methods to predict post-traumatic stress disorder in stroke patients in China
Ying Li1, Chuang Pan2, Yue Gu3
1College of Sports Science, Jishou University, Jishou, Hunan, China.
Frontiers in Psychiatry
|December 8, 2025
Summary
Machine learning models can predict post-stroke Post-Traumatic Stress Disorder (PTSD). The Random Forest model best identified risk factors like stroke type and sleep quality for PTSD in stroke survivors.
Area of Science:
- Neurology
- Psychiatry
- Data Science
Background:
- Post-stroke Post-Traumatic Stress Disorder (PTSD) affects a significant portion of stroke survivors.
- Identifying risk factors and developing predictive models are crucial for early intervention.
Purpose of the Study:
- To develop and validate machine learning models for predicting post-stroke PTSD.
- To identify key risk factors associated with PTSD development after stroke.
Main Methods:
- Employed Logistic Regression, Random Forest (RF), and K-nearest neighbor algorithms.
- A cohort of 249 stroke patients was divided into training and validation groups.
- Model performance was evaluated to select the optimal predictive algorithm.
Main Results:
- The incidence of PTSD in stroke patients was 40.56%.
- The Random Forest (RF) model demonstrated the best predictive performance.
- Key risk factors identified include stroke type, sleep quality, hospitalization method, income, hypertension, gender, marital status, exercise, and education.
Conclusions:
- The RF-based model offers superior predictive capability for post-stroke PTSD.
- Findings aid clinicians in identifying high-risk individuals for targeted preventive strategies.
- Understanding risk factors like stroke type and sleep is vital for PTSD prevention.
