NeuroDx-LM: A Clinical Large-Scale Model for EEG-based Neurological Disorder Detection

TL;DR

NeuroDx-LM is a large-scale EEG model with novel embedding and training strategies that improves neurological disorder detection, demonstrating state-of-the-art results on seizure and schizophrenia datasets.

Contribution

The paper introduces NeuroDx-LM, a new large-scale EEG model with adaptive embedding and progressive training for better clinical disorder detection.

Findings

Achieved state-of-the-art performance on CHB-MIT seizure dataset.

Achieved state-of-the-art performance on schizophrenia detection dataset.

Demonstrated potential for clinical application of EEG-based models.

Abstract

Large-scale models pre-trained on Electroencephalography (EEG) have shown promise in clinical applications such as neurological disorder detection. However, the practical deployment of EEG-based large-scale models faces critical challenges such as limited labeled EEG data and suboptimal performance in clinical scenarios. To address these issues, we propose NeuroDx-LM, a novel large-scale model specifically designed for detecting EEG-based neurological disorders. Our key contributions include (i) a Selective Temporal-Frequency Embedding mechanism that adaptively captures complex temporal and spectral patterns in EEG signals; and (ii) a Progressive Feature-Aware Training strategy that refines feature representation in a two-stage process. In the first stage, our model learns the fundamental discriminative features of EEG activities; in the second stage, the model further extracts more specialized fine-grained features for accurate diagnostic performance. We evaluated NeuroDx-LM on the CHB-MIT and Schizophrenia datasets, achieving state-of-the-art performance in EEG-based seizure and schizophrenia detection, respectively. These results demonstrate the great potential of EEG-based large-scale models to advance clinical applicability. Our code is available at https://github.com/LetItBe12345/NeuroDx-LM.