SOMAS reliably quantifies muscle tone and movements during sleep from EDF+ files using open-source algorithms, with the potential of enhancing reproducibility and collaboration in research on sleep-related movement disorders.
SOMAS-derived atonia index and leg movement indices strongly correlated with those from Hypnolab, a non-open access software.
The Distribution of Normalized EMG values (DNE) and atonia index in REM sleep effectively differentiated patients with isolated REM sleep behavior disorder (iRBD) from other patients and controls.
Periodic leg movement and periodicity indices differentiated patients with restless legs syndrome (RLS) from controls.
SOMAS quantifies muscle tone using two complementary metrics: the atonia index and the distribution of normalized electromyography values (DNE).
SOMAS is an open-source software system that fills a gap in existing sleep analysis tools by providing information on both muscle tone and movements.
This research describes the development and validation of SOMAS (Sleep Open-source Muscle Activity Analysis System), a freely available software tool for analyzing muscle activity during sleep. The software can measure both muscle tone — how relaxed or tense muscles are during different sleep stages — and leg movements during sleep, two important aspects of diagnosing sleep-related movement disorders. Unlike existing tools that are proprietary or closed-source, SOMAS works with standard sleep recording files and does not require sharing patient data online, making it more accessible and privacy-friendly. The researchers tested SOMAS on recordings from 25 individuals, including patients with isolated REM sleep behavior disorder (iRBD, a condition where people physically act out dreams), restless legs syndrome (RLS), sleep breathing disorders, and healthy controls. They found that SOMAS produced results nearly identical to those from an established but non-open-source software called Hypnolab, with correlation coefficients above 0.97. SOMAS was also able to distinguish iRBD patients from other groups with very high accuracy (AUC 0.89–1.00) and could differentiate RLS patients from controls with moderate accuracy (AUC 0.71–0.75). This research suggests that SOMAS could be a valuable tool for sleep researchers worldwide, particularly because its open-source nature means that the algorithms are transparent and reproducible. This is important for scientific research, where independent verification of results is a cornerstone of reliability. By making such a tool freely available, the authors aim to improve consistency across different research groups studying sleep disorders involving abnormal muscle activity.
Cesari M, Ferri R, Högl B, Stefani A, Silvani A. (2026). SOMAS - an open-source software for the analysis of muscle activity during sleep.. Sleep medicine. https://doi.org/10.1016/j.sleep.2026.108791