Dynamic cross-modal spatio-temporal graph attention network: Multimodal coupling analysis in sleep stage classification

Key Points

• Sleep stage classification improved through a dynamic approach, enhancing accuracy across various conditions.
• Key performance metric shows a higher accuracy rate of 85% when utilizing the proposed graph attention network.
• Analysis involves the application of a spatio-temporal graph attention model to handle multiple data modalities effectively.
• Findings highlight the potential for more precise sleep monitoring tools, benefiting those with sleep disorders.