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

TL;DR

This paper introduces a pre-trained transformer model trained on long-term deep brain stimulation data to decode Parkinson's symptoms without needing patient-specific training, leveraging an extended context window and a novel loss function.

Contribution

It presents a novel foundation model for neural electrophysiology that enables symptom decoding across patients without individual training, using a specialized loss function and long-term data.

Findings

Effective symptom decoding without patient-specific training

Extended 30-minute context window improves decoding accuracy

Novel loss function corrects frequency bias in electrophysiological data

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.