Pith. sign in

REVIEW 7 cited by

SleepFM: Multi-modal Representation Learning for Sleep Across Brain Activity, ECG and Respiratory Signals

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2405.17766 v1 pith:2EHGQJT6 submitted 2024-05-28 cs.LG cs.AIeess.SP

classification cs.LGcs.AIeess.SP
keywords sleepmulti-modalsleepfmlearningauprcaurocbraincontrastive
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Sleep is a complex physiological process evaluated through various modalities recording electrical brain, cardiac, and respiratory activities. We curate a large polysomnography dataset from over 14,000 participants comprising over 100,000 hours of multi-modal sleep recordings. Leveraging this extensive dataset, we developed SleepFM, the first multi-modal foundation model for sleep analysis. We show that a novel leave-one-out approach for contrastive learning significantly improves downstream task performance compared to representations from standard pairwise contrastive learning. A logistic regression model trained on SleepFM's learned embeddings outperforms an end-to-end trained convolutional neural network (CNN) on sleep stage classification (macro AUROC 0.88 vs 0.72 and macro AUPRC 0.72 vs 0.48) and sleep disordered breathing detection (AUROC 0.85 vs 0.69 and AUPRC 0.77 vs 0.61). Notably, the learned embeddings achieve 48% top-1 average accuracy in retrieving the corresponding recording clips of other modalities from 90,000 candidates. This work demonstrates the value of holistic multi-modal sleep modeling to fully capture the richness of sleep recordings. SleepFM is open source and available at https://github.com/rthapa84/sleepfm-codebase.

Discussion (0). Sign in to comment.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. STEAM: A Spatio-TEmporal Alignment Mixture-of-Experts Model with Hierarchical Pre-training for EEG Decoding

    cs.CV 2026-08 conditional novelty 6.0 of 10

    STEAM, a dual-branch spatio-temporal mixture-of-experts EEG model with two-stage pre-training, reports the best average rank across seven downstream EEG datasets and fourteen evaluation settings.

  2. Agentic AI-enabled discovery across large-scale sleep physiology

    cs.MA 2026-07 conditional novelty 6.0 of 10

    A human-guided agentic AI pipeline analyzed ~124,000 polysomnograms and reported five sleep findings, including an association between reduced N2 brain coupling and incident Parkinson's (HR 1.48) and Alzheimer's (HR 1...

  3. OSF: On Pre-training and Scaling of Sleep Foundation Models

    cs.LG 2026-02 conditional novelty 6.0 of 10

    Channel-masked self-supervised pretraining on a 166,500-hour multi-source sleep corpus yields OSF, which generalizes better to missing channels and scales with data and model size.

  4. SleepMaMi: A Universal Sleep Foundation Model for Integrating Macro- and Micro-structures

    cs.AI 2026-02 conditional novelty 6.0 of 10

    SleepMaMi, a dual-encoder sleep foundation model pretrained on 158K hours of PSG, matches or beats existing sleep foundation models on staging, apnea segmentation, and disease prediction.

  5. PiCME: Pipeline for Contrastive Modality Evaluation and Encoding in the MIMIC Dataset

    cs.LG 2025-07 conditional novelty 6.0 of 10

    PiCME shows contrastive learning peaks at three modalities in MIMIC, and a Modality-Gated LSTM with contrastively learned weights improves five-modality mortality prediction over supervised baselines.

  6. Self-DANA: A Resource-Efficient Channel-Adaptive Self-Supervised Approach for ECG Foundation Models

    eess.SP 2025-07 conditional novelty 6.0 of 10

    Self-DANA combines dimension-adaptive pooling with random lead selection to fine-tune ECG foundation models on reduced-lead inputs, cutting memory and time while maintaining diagnostic accuracy.

  7. Estimating Markers of Driving Stress through Multimodal Physiological Monitoring

    eess.SP 2025-07 conditional novelty 4.0 of 10

    A multimodal physiological classifier distinguishes stressor periods from free driving in a simulator with AUROC 0.812, but the per-second real-time claim is weakened by a centered, non-causal window.

Pith tools