FemWear is a parameter‑efficient wearable foundation model specifically tailored for women's health. It repurposes a pretrained multimodal wearable backbone by training only 239,236 encoder parameters—just 1.11% of the original 21.54M—using low‑rank residual adapters and causal task‑family heads to create a shared longitudinal representation for menstrual, symptom, affective, sleep/recovery, autonomic, activity, and pregnancy outcomes. Evaluations across six cohorts and 63 metrics show improvements in cycle‑phase macro‑F1 and reductions in mean absolute error for cramps, mood symptoms, and sleep problems, while maintaining the OpenMHC ability‑retention benchmark.
By Yifan Wang, Chenzhong Li
arXiv:2608. 08244v1 Announce Type: new Abstract: General wearable foundation models are pretrained across broad sensor streams and populations, but are not designed around women's-health tasks.
By Yifan Wang, Chenzhong Li
arXiv:2608. 05982v1 Announce Type: new Abstract: Inadequate target--disease linkage accounts for 40--50\% of Phase~II efficacy failures, so anticipating which programmes will advance would let sponsors back the hypotheses most likely to reach patients.
By Pui Chung Siu, Claudia Cabrera, Mani Mudaliar, Arkaitz Zubiaga
DyssomniaNN is a multi‑scale computational model that integrates Borbely’s two‑process sleep regulation, Saper’s hypothalamic flip‑flop circuit, and a corticothalamic spiking neural network to simulate human sleep disorders. The model reproduces empirical EEG spectra across five clinical cohorts—healthy control, delayed‑phase circadian rhythm disorder, chronic insomnia, paradoxical insomnia, and narcolepsy type 1—within a 72‑hour photoperiod. Introducing the Well‑Rested Index, the simulations show that total sleep time does not reliably predict restorative recovery, as exemplified by paradoxical insomnia’s normal sleep duration but low restoration and narcolepsy’s severe nocturnal fragmentation.
By Daniel Szelogowski
arXiv:2606. 00496v1 Announce Type: new Abstract: Oscillatory neural signals such as electroencephalography (EEG) and local field potentials (LFPs) show phase relationships that coordinate communication across brain regions.
By Jack Goffinet, Casey Hanks, David E. Carlson
arXiv:2607. 00431v1 Announce Type: new Abstract: Forecasting models for health-signal digital twins must preserve the oscillatory, frequency, phase, and state-transition dynamics of physiological signals, yet the pointwise metrics used to benchmark them cannot detect when these fundamental properties are lost.
By Md Rakibul Haque, Shireen Elhabian, Warren Woodrich Pettine