From data to decisions: AI and functional connectivity for diagnosis, prognosis, and recovery prediction in stroke

Abstract

Stroke is a severe medical condition which may lead to permanent disability conditions. The initial 8 weeks following a stroke are crucial for rehabilitation, as most recovery occurs during this period. Personalized approaches and predictive biomarkers are needed for tailored rehabilitation. In this context, EEG brain connectivity and Artificial Intelligence (AI) can play a crucial role in diagnosing and predicting stroke outcomes efficiently. In the present study, 127 patients with subacute ischemic lesions and 90 age- and gender-matched healthy controls were enrolled. EEG recordings were obtained from each participant within 15 days of stroke onset. Clinical evaluations were performed at baseline and at 40-days follow-up using the National Institutes of Health Stroke Scale (NIHSS). Functional connectivity analysis was conducted using Total Coherence (TotCoh) and Small Word (SW). Quadratic support vector machines (SVM) algorithms were implemented to classify healthy subjects compared to stroke patients (Healthy vs Stroke), determine the affected hemisphere (Left vs Right Hemisphere), and predict functional recovery (Functional Recovery Prediction). In the classification for Functional Recovery Prediction, an accuracy of 94.75%, sensitivity of 96.27% specificity of 92.33%, and AUC of 0.95 were achieved; for Healthy vs Stroke, an accuracy of 99.09%, sensitivity of 100%, specificity of 98.46%, and AUC of 0.99 were achieved. For Left vs Right Hemisphere classification, accuracy was 86.77%, sensitivity was 91.44%, specificity was 80.33%, and AUC was 0.87. These findings highlight the potential of utilizing functional connectivity measures based on EEG in combination with AI algorithms to improve patient outcomes by targeted rehabilitation interventions.

Keywords: AI; Coherence; Connectivity; EEG; Graph; Neurorehabilitation; Prediction; Stroke.

Fig. 1

Classification pipeline: the protocol involves several steps utilizing EEG data to derive functional connectivity measures (TotCoh and SW). Initially, the presence or absence of stroke is identified using the connectivity measures derived from the EEG data. Using the same connectivity measures, the protocol proceeds to identify the location of the lesion, distinguishing between the right hemisphere and the left hemisphere. Following, utilizing the NIHSS and connectivity data at T₀, a prediction is made regarding the patient’s functional recovery at the follow-up evaluation after approximately 40 days (T₁). The prediction involves determining whether the patient will recover more than 4 points (+ Outcome) or fewer than 4 points (- Outcome) on the NIHSS scale

Fig. 2

Receiver operating xharacteristic (ROC) curves and their corresponding 95% confidence interval (CI) for the three distinct classification problems (Healthy vs Stroke, Right vs Left Hemisphere, and Functional Recovery Prediction) and for both datasets employed in the study (one comprising all available features, and the other containing only the features selected through a feature selection technique). The first row corresponds to the dataset with all features, while the second row represents the dataset with selected features