Sleep Stage Detection from Actigraphy and Heart Rate Using an Attention-Based Model

Amyn Sorkhilalehloo, Nasim Montazeri Ghahjaverestan

Confirmation: I have read and agree with the IEEE BHI 2025 conference submission's policy on behalf of myself and my co-authors.

Keywords: Sleep, Sleep Stage, Actigraphy, Heart Rate, Deep Learning

TL;DR: This work presents a subject-specific attention-based deep model for detecting sleep stages using heart rate and actigraphy signals.

Abstract:

Sleep plays a crucial role in human well-
being, while insufficient sleep affects cognitive function,
decision-making, and overall health. Sleep assessment
via polysomnography (PSG) is time-consuming, resource-
intensive, and limited to in-laboratory sleep testing. To
address the challenges of PSG, wearable sleep screen-
ing devices have been widely used, especially to detect
wakefulness and sleep stages. This study proposes deep
models for the detection of wakefulness versus differ-
ent stages of sleep using heart rate and wrist actigraphy
extracted from the multi-ethnic study of atherosclerosis
(MESA) sleep dataset. First, two sets of features were ex-
tracted from heart rate and actigraphy, which were sepa-
rately fed into two separate branches of convolution neural
network (CNN), then merged and fed to a deep classifier.
The model detected wakefulness versus sleep and different
sleep stages with the accuracies of 88.19% and 79.6%
respectively. This work showed that combining heart rate,
actigraphy signals, and demographic data in a deep frame-
work could improve sleep stage-staging performance. This
study offers a subject-specific approach for sleep assess-
ment based on convenient wearables.

Track: 1. Biomedical Sensor Informatics

Registration Id: S5NLPMZMZ7T

Submission Number: 129