Physiology as Language: Translating Respiration to EEG during Sleep

This paper introduces a novel cross-physiology translation task: synthesizing sleep electroencephalography (EEG) from respiration signals. To address the significant complexity gap between the two modalities, we propose a waveform-conditional generative framework that preserves fine-grained respiratory dynamics while constraining the EEG target space through discrete tokenization. Trained on over 28,000 individuals, our model achieves a 7% Mean Absolute Error in EEG spectrogram reconstruction. Beyond reconstruction, the synthesized EEG supports downstream tasks with performance comparable to ground truth EEG on age estimation (MAE 5.0 vs. 5.1 years), sex detection (AUROC 0.81 vs. 0.82), and sleep staging (Accuracy 0.84 vs. 0.88), significantly outperforming baselines trained directly on breathing. Finally, we demonstrate that the framework generalizes to contactless sensing by synthesizing EEG from wireless radio-frequency reflections, highlighting the feasibility of remote, non-contact neurological assessment during sleep.

Sleep EEG is one of the most important tools for measuring brain activity and diagnosing neurological and sleep disorders, but collecting EEG requires uncomfortable head sensors and specialized sleep studies. In contrast, breathing signals are much easier to measure using wearable devices or even contactless wireless sensors. Our research asks whether it is possible to reconstruct meaningful brain activity during sleep using breathing alone. We developed a machine learning system that translates breathing patterns into sleep EEG signals. The model learns relationships between respiratory rhythms and brain activity by training on more than 33,000 nights of sleep data from over 28,000 people. The generated EEG closely matches real EEG recordings and preserves important sleep features such as slow waves and sleep spindles. We further show that the synthesized EEG can support practical clinical tasks including sleep staging, age estimation, and sex classification at performance levels close to real EEG. Finally, we demonstrate that EEG can even be inferred from contactless wireless reflections, opening the possibility of comfortable, remote neurological monitoring during sleep without head-worn sensors.