Multimodal Sleep Stage and Sleep Apnea Classification Using Vision Transformer: A Multitask Explainable Learning Approach

TL;DR

This paper introduces a multimodal, multitask Vision Transformer model that simultaneously classifies sleep stages and sleep apnea, leveraging multimodal signals and explainable attention mechanisms to improve accuracy and interpretability.

Contribution

The study presents a novel 1D-Vision Transformer framework for concurrent sleep stage and sleep disorder classification using multimodal data, with explainable attention analysis.

Findings

Achieved 78% accuracy for sleep stage classification

Achieved 74% accuracy for sleep apnea detection

Attention analysis highlights respiratory features' importance

Abstract

Sleep is an essential component of human physiology, contributing significantly to overall health and quality of life. Accurate sleep staging and disorder detection are crucial for assessing sleep quality. Studies in the literature have proposed PSG-based approaches and machine-learning methods utilizing single-modality signals. However, existing methods often lack multimodal, multilabel frameworks and address sleep stages and disorders classification separately. In this paper, we propose a 1D-Vision Transformer for simultaneous classification of sleep stages and sleep disorders. Our method exploits the sleep disorders' correlation with specific sleep stage patterns and performs a simultaneous identification of a sleep stage and sleep disorder. The model is trained and tested using multimodal-multilabel sensory data (including photoplethysmogram, respiratory flow, and respiratory effort signals). The proposed method shows an overall accuracy (cohen's Kappa) of 78% (0.66) for five-stage sleep classification and 74% (0.58) for sleep apnea classification. Moreover, we analyzed the encoder attention weights to clarify our models' predictions and investigate the influence different features have on the models' outputs. The result shows that identified patterns, such as respiratory troughs and peaks, make a higher contribution to the final classification process.