What is the clinical evidence from this study?

Study design: Other. Population: Ischemic stroke (n=64). Intervention: Multi-modal perceptron model integrating respiratory signals vs. State-of-the-art machine learning models and models without respiratory signals. Primary outcome: Prediction accuracy.

What does this research mean for the field?

A multi-modal stroke-risk prediction model achieved 84.91% accuracy, identifying CO2-derived features as critical predictors alongside ECG, EMG, and blood pressure signals. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.CHALLENGES_CONSENSUS.

What question did this study set out to answer?

The aim is to develop a reliable stroke-risk prediction model using multiple physiological signals, including underutilized respiratory data.

February 5, 2026Scientific Reports0 citationsOpen Access

Explainable multi-modal approach for uncovering key predictors of stroke-risk from ECG, EMG, blood pressure, and respiratory signals

JKJalal KrayemMonash University Malaysia LWLily WongMonash University Malaysia LTLai Kuan ThamUniversity of Malaya

Key Result

A single perceptron model integrating respiratory signals alongside blood pressure, ECG, and EMG achieved a prediction accuracy of 84.91% for stroke risk.

Key Points

The aim is to develop a reliable stroke-risk prediction model using multiple physiological signals, including underutilized respiratory data.
Integrated respiratory signals with ECG, EMG, and blood pressure for model development.
Utilized a single perceptron model for prediction and evaluated with a dataset of 64 subjects.
Applied explainable AI techniques like LIME, SHAP, and Anchors to interpret model decisions.
Achieved a prediction accuracy of 84.91%, outperforming existing ML and DL methods.
SHAP identified CO2-derived features as critical predictors for stroke risk.
Findings highlight the importance of integrating respiratory signals in stroke-risk assessment.

Structured PICO

Population

Dataset of 64 subjects

Intervention

Explainable multi-modal stroke-risk prediction model (single perceptron) integrating respiratory signals (carbon dioxide and respiration flow), blood pressure, ECG, and EMG

Comparator

State-of-the-art machine learning (ML) and deep learning (DL) methods

Outcome

Stroke-risk prediction accuracy

An explainable multi-modal AI model integrating respiratory signals with ECG, EMG, and BP achieved 84.91% accuracy in stroke-risk prediction, highlighting the overlooked value of CO2-derived features.

Limitations

Small sample size of 64 subjects after excluding those with missing signals or incomplete tests
Reliance on post-hoc explainers which do not faithfully represent the true internal mechanisms of the model

Abstract

Accurate and timely stroke-risk prediction is necessary to help patients at risk take guided measures, as stroke remains a leading cause of death and long-term disability worldwide. While there are several developed stroke-risk prediction models using various bio-signals, it remains unclear which signal or signal feature carries the most information towards stroke. Additionally, respiratory signals have not been included in these models, despite research showing their relation to stroke disease. We address these gaps by developing an explainable multi-modal stroke-risk prediction model that integrates respiratory signals (carbon dioxide (COFormula: see text), and respiration flow) alongside blood pressure (BP), electrocardiogram (ECG) and electromyography (EMG). We developed a single perceptron model which achieved a prediction accuracy of 84.91% on a dataset of 64 subjects, outperforming state-of-the-art machine learning (ML) and deep learning (DL) methods. Explainable Artificial Intelligence (XAI) techniques, namely LIME, SHAP, and Anchors, were applied to interpret the model's decisions. The Model Explanation Metric Consensus (MEMC) XAI evaluation metric revealed SHAP as the most reliable explainer, which identified COFormula: see text-derived features as critical predictors. The findings demonstrate the overlooked value of respiratory signals in stroke-risk prediction, as well as the importance of explainable multi-modal approaches in advancing stroke-risk prediction, and enabling clinicians to better understand and trust AI models for improved patient care.

AI से पूछें

Bookmark

View Full Paper