Sleep-Energy: An Energy Optimization Method to Sleep Stage Scoring

Sleep is essential for physical and mental health. Polysomnography (PSG) procedures are labour-intensive time-consuming, making diagnosing sleep disorders difficult. Automatic staging using ML-based methods has been studied extensively, but frequently provides noisier predictions incompatible with typical manually annotated hypnograms. We propose an energy optimization method to improve the quality of hypnograms generated by automatic procedures. The evaluates system’s total based on conditional probabilities each epoch’s stage employs minimisation procedure. It can be used as a meta-optimisation layer over sequences any classifier that generates prediction probabilities. Results show it improves accuracy state-of-the-art Deep Learning models two public datasets.