English

Pre-trained Transformer-models using chronic invasive electrophysiology for symptom decoding without patient-individual training

Human-Computer Interaction 2025-08-15 v1 Machine Learning

Abstract

Neural decoding of pathological and physiological states can enable patient-individualized closed-loop neuromodulation therapy. Recent advances in pre-trained large-scale foundation models offer the potential for generalized state estimation without patient-individual training. Here we present a foundation model trained on chronic longitudinal deep brain stimulation recordings spanning over 24 days. Adhering to long time-scale symptom fluctuations, we highlight the extended context window of 30 minutes. We present an optimized pre-training loss function for neural electrophysiological data that corrects for the frequency bias of common masked auto-encoder loss functions due to the 1-over-f power law. We show in a downstream task the decoding of Parkinson's disease symptoms with leave-one-subject-out cross-validation without patient-individual training.

Cite

@article{arxiv.2508.10160,
  title  = {Pre-trained Transformer-models using chronic invasive electrophysiology for symptom decoding without patient-individual training},
  author = {Timon Merk and Saeed Salehi and Richard M. Koehler and Qiming Cui and Maria Olaru and Amelia Hahn and Nicole R. Provenza and Simon Little and Reza Abbasi-Asl and Phil A. Starr and Wolf-Julian Neumann},
  journal= {arXiv preprint arXiv:2508.10160},
  year   = {2025}
}

Comments

5 pages, 6 figures

R2 v1 2026-07-01T04:48:51.977Z