Research map / Large language models / Language, safety and interpretability
Physiology as Language: Translating Respiration to EEG during Sleep
Kaiwen Zha, Chao Li, Hao He, Peng Cao, Tianhong Li, Ali Mirzazadeh, Ellen Zhang, Jong Woo Lee, Yoon Kim, Dina Katabi
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| synthesize EEG from breathing before predicting age | Resp-to-EEG translation · Age estimation | MAE (years): 7.66 -> 5.14 | Improved |
| synthesize EEG from breathing before predicting sex | Resp-to-EEG translation · Sex classification | AUROC: 0.768 -> 0.814 | Improved |
| synthesize EEG from breathing before sleep staging | Resp-to-EEG translation · Sleep staging | accuracy: 0.812 -> 0.839 | Improved |
| use raw linear breathing embedding instead of heavy encoders or time-frequency conversion (ablation) | Resp-to-EEG translation · EEG reconstruction | MAE | Improved |
| use masked generation instead of autoregressive generation (ablation) | Resp-to-EEG translation · EEG reconstruction | reconstruction fidelity | Improved |
| use RF breathing instead of breathing belt | Resp-to-EEG translation · MGH EEG reconstruction | MAE: 0.069 ± 0.026 -> 0.076 ± 0.027 | Mixed |
| use RF breathing instead of belt for age estimation | Resp-to-EEG translation · MGH | MAE (years): 7.96 -> 8.83 | Hurt |
| use synthesized EEG instead of stage-average baseline | Resp-to-EEG translation · band-wise correlation | Pearson correlation: +14%, +19%, +28%, +22%, +8% (Delta to Beta) | Improved |
Closest papers by meaning
- SleepFM: Multi-modal Representation Learning for Sleep Across Brain Activity, ECG and Respiratory Signals (ICML 2024 · similarity 0.85)
- LEAF: Language-EEG Aligned Foundation Model for Brain-Computer Interfaces (NeurIPS 2026 · similarity 0.85)
- SENSE: Semantic Neural Speech Synthesis from Brain Dynamics via Spatial Graph Encoding (NeurIPS 2026 · similarity 0.85)
- CBraMod: A Criss-Cross Brain Foundation Model for EEG Decoding (ICLR 2025 · similarity 0.84)
- Robust Sleep Staging over Incomplete Multimodal Physiological Signals via Contrastive Imagination (NeurIPS 2024 · similarity 0.84)
- What do EEG Foundation Models Capture from Human Brain Signals? (NeurIPS 2026 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
