Key result
A deep learning CNN-LSTM model using raw ECG and PPG bio-signals collected during walking achieved a stroke prediction accuracy of 99.15% in elderly patients.
Why the study?
Existing imaging methods for detecting stroke precursor symptoms are costly, time-consuming, and difficult to deploy for real-time early diagnosis.
Population
Elderly individuals
Comparison
CNN-LSTM model vs RandomForest and C4.5 decision tree models
Design
Machine learning model development and validation study
Authors
Loading...
May aid real-time stroke precursor detection; leaves open validation in larger prospective studies.
An ensemble deep learning model using real-time ECG and PPG bio-signals demonstrated high accuracy (>99%) in predicting stroke prognostic symptoms in the elderly.
Yu et al. (2022) studied Stroke (n=574). AI-based stroke prediction model (CNN-LSTM) using ECG and PPG bio-signals vs. Other machine learning models (e.g., Random Forest, C4.5) was evaluated on Prediction accuracy of stroke disease. A deep learning CNN-LSTM model using raw ECG and PPG bio-signals collected during walking achieved a stroke prediction accuracy of 99.15% in elderly patients.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: