Salvato in:
| Autori principali: | , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2508.10160 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909736611348480 |
|---|---|
| author | Merk, Timon Salehi, Saeed Koehler, Richard M. Cui, Qiming Olaru, Maria Hahn, Amelia Provenza, Nicole R. Little, Simon Abbasi-Asl, Reza Starr, Phil A. Neumann, Wolf-Julian |
| author_facet | Merk, Timon Salehi, Saeed Koehler, Richard M. Cui, Qiming Olaru, Maria Hahn, Amelia Provenza, Nicole R. Little, Simon Abbasi-Asl, Reza Starr, Phil A. Neumann, Wolf-Julian |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_10160 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Pre-trained Transformer-models using chronic invasive electrophysiology for symptom decoding without patient-individual training Merk, Timon Salehi, Saeed Koehler, Richard M. Cui, Qiming Olaru, Maria Hahn, Amelia Provenza, Nicole R. Little, Simon Abbasi-Asl, Reza Starr, Phil A. Neumann, Wolf-Julian Human-Computer Interaction Machine Learning 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. |
| title | Pre-trained Transformer-models using chronic invasive electrophysiology for symptom decoding without patient-individual training |
| topic | Human-Computer Interaction Machine Learning |
| url | https://arxiv.org/abs/2508.10160 |