AI Generalisation Gap In Comorbid Sleep Disorder Staging

TL;DR

This paper examines the limitations of deep learning models in sleep staging for stroke patients, introduces a new dataset, and emphasizes the need for disease-specific models with clinical validation.

Contribution

It introduces iSLEEPS, a new clinical stroke sleep dataset, and evaluates model generalization issues, highlighting the importance of subject-aware approaches.

Findings

Cross-domain performance between healthy and diseased subjects is poor.

Attention visualizations show models focus on uninformative EEG regions in patients.

Significant sleep architecture differences highlight need for disease-specific models.

Abstract

Accurate sleep staging is essential for diagnosing OSA and hypopnea in stroke patients. Although PSG is reliable, it is costly, labor-intensive, and manually scored. While deep learning enables automated EEG-based sleep staging in healthy subjects, our analysis shows poor generalization to clinical populations with disrupted sleep. Using Grad-CAM interpretations, we systematically demonstrate this limitation. We introduce iSLEEPS, a newly clinically annotated ischemic stroke dataset (to be publicly released), and evaluate a SE-ResNet plus bidirectional LSTM model for single-channel EEG sleep staging. As expected, cross-domain performance between healthy and diseased subjects is poor. Attention visualizations, supported by clinical expert feedback, show the model focuses on physiologically uninformative EEG regions in patient data. Statistical and computational analyses further confirm significant sleep architecture differences between healthy and ischemic stroke cohorts, highlighting the need for subject-aware or disease-specific models with clinical validation before deployment. A summary of the paper and the code is available at https://himalayansaswatabose.github.io/iSLEEPS_Explainability.github.io/