Back

Vision-Based Automated Severity Rating of REM Sleep Behavior Disorder: From Heuristic Features to Foundation Models

Hwang, J.; van Pesch, N.; Raval, B.; Abdelfattah, M.; Gafsi, A.; Bertolaso, A.; Ryu, K. H.; Marwaha, S.; Sum-Ping, O.; Cesari, M.; Stefani, A.; Brink-Kjaer, A.; Mignot, E.; Alahi, A.; During, E.

2026-07-04 neurology
10.64898/2026.07.02.26356954 medRxiv
Show abstract

Automated severity rating of rapid eye movement (REM) sleep behavior disorder (RBD) movements would enable multi-night monitoring to detect potentially injurious behaviors and provide objective endpoints for clinical trials. We compared a heuristic classifier using optical flow-derived features against V-JEPA2, a self-supervised video foundation model, for clip-level severity classification (3329 mild versus 284 moderate-to-severe) of in-laboratory video-polysomnography infrared recordings in 86 isolated RBD patients. V-JEPA2 with checkpoint fine-tuning and maximum optical flow-based frame sampling achieved the best performance across both evaluation conditions -- Macro F1 of 0.76 and 93% accuracy in the clip-level split, and 0.68 and 85% in the patient-level split -- outperforming heuristic and domain-specific pretrained models. Clip duration was the dominant heuristic predictor. Whole-night severity scores preserved patient-level ordering despite systematic overestimation, with V-JEPA2 achieving a mean absolute error of 25% versus 52% for the heuristic classifier. These findings establish a foundation for objective, home-deployable monitoring of RBD severity.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

1
npj Digital Medicine
118 papers in training set
Top 0.1%
38.8%
2
Sleep
58 papers in training set
Top 0.2%
6.2%
3
Communications Medicine
113 papers in training set
Top 0.4%
5.4%
50% of probability mass above
4
Nature Medicine
125 papers in training set
Top 0.3%
5.4%
5
IEEE Access
35 papers in training set
Top 0.2%
5.1%
6
Journal of Sleep Research
36 papers in training set
Top 0.2%
5.1%
7
Nature Communications
5641 papers in training set
Top 33%
4.0%
8
Scientific Reports
3612 papers in training set
Top 39%
2.7%
9
PLOS ONE
5266 papers in training set
Top 41%
2.6%
10
Frontiers in Neuroscience
256 papers in training set
Top 2%
2.4%
11
eLife
5828 papers in training set
Top 58%
1.1%
12
PLOS Digital Health
106 papers in training set
Top 3%
1.1%
13
PLOS Computational Biology
1863 papers in training set
Top 17%
1.1%
14
npj Parkinson's Disease
105 papers in training set
Top 1%
1.0%
15
eBioMedicine
183 papers in training set
Top 5%
1.0%
16
Frontiers in Neurology
102 papers in training set
Top 3%
0.8%
17
Annals of Neurology
64 papers in training set
Top 2%
0.8%
18
Movement Disorders
71 papers in training set
Top 0.8%
0.8%
19
Sleep Medicine
19 papers in training set
Top 0.3%
0.8%
20
eClinicalMedicine
77 papers in training set
Top 3%
0.6%
21
Frontiers in Neuroinformatics
41 papers in training set
Top 0.8%
0.6%
22
Journal of Medical Internet Research
87 papers in training set
Top 3%
0.6%
23
Journal of Biological Rhythms
25 papers in training set
Top 0.3%
0.6%
24
Neuroscience of Consciousness
16 papers in training set
Top 0.4%
0.6%
25
JMIR mHealth and uHealth
11 papers in training set
Top 0.4%
0.6%