PRISM: Exploring Heterogeneous Pretrained EEG Foundation Model Transfer to Clinical Differential Diagnosis
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908861166780416 |
|---|---|
| author | Lahiri, Jeet Bandhu Runwal, Parshva Kulkarni, Arvasu Jain, Mahir Mishra, Aditya Ray Panwar, Siddharth Singh, Sandeep |
| author_facet | Lahiri, Jeet Bandhu Runwal, Parshva Kulkarni, Arvasu Jain, Mahir Mishra, Aditya Ray Panwar, Siddharth Singh, Sandeep |
| contents | EEG foundation models are typically pretrained on narrow-source clinical archives and evaluated on benchmarks from the same ecosystem, leaving unclear whether representations encode neural physiology or recording-distribution artifacts. We introduce PRISM (Population Representative Invariant Signal Model), a masked autoencoder ablated along two axes -- pretraining population and downstream adaptation -- with architecture and preprocessing fixed. We compare a narrow-source EU/US corpus (TUH + PhysioNet) against a geographically diverse pool augmented with multi-center South Asian clinical recordings across multiple EEG systems. Three findings emerge. First, narrow-source pretraining yields stronger linear probes on distribution-matched benchmarks, while diverse pretraining produces more adaptable representations under fine-tuning -- a trade-off invisible under single-protocol evaluation. Trained on three source corpora, PRISM matches or outperforms REVE (92 datasets, 60,000+ hours) on the majority of tasks, demonstrating that targeted diversity can substitute for indiscriminate scale and that dataset count is a confounding variable in model comparison. Second, on a clinically challenging and previously untested task -- distinguishing epilepsy from diagnostic mimickers via interictal EEG -- the diverse checkpoint outperforms the narrow-source checkpoint by +12.3 pp balanced accuracy, the largest gap across all evaluations. Third, systematic inconsistencies between EEG-Bench and EEG-FM-Bench reverse model rankings on identical datasets by up to 24 pp; we identify six concrete sources including split construction, checkpoint selection, segment length, and normalization, showing these factors compound non-additively. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_02268 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PRISM: Exploring Heterogeneous Pretrained EEG Foundation Model Transfer to Clinical Differential Diagnosis Lahiri, Jeet Bandhu Runwal, Parshva Kulkarni, Arvasu Jain, Mahir Mishra, Aditya Ray Panwar, Siddharth Singh, Sandeep Machine Learning Artificial Intelligence EEG foundation models are typically pretrained on narrow-source clinical archives and evaluated on benchmarks from the same ecosystem, leaving unclear whether representations encode neural physiology or recording-distribution artifacts. We introduce PRISM (Population Representative Invariant Signal Model), a masked autoencoder ablated along two axes -- pretraining population and downstream adaptation -- with architecture and preprocessing fixed. We compare a narrow-source EU/US corpus (TUH + PhysioNet) against a geographically diverse pool augmented with multi-center South Asian clinical recordings across multiple EEG systems. Three findings emerge. First, narrow-source pretraining yields stronger linear probes on distribution-matched benchmarks, while diverse pretraining produces more adaptable representations under fine-tuning -- a trade-off invisible under single-protocol evaluation. Trained on three source corpora, PRISM matches or outperforms REVE (92 datasets, 60,000+ hours) on the majority of tasks, demonstrating that targeted diversity can substitute for indiscriminate scale and that dataset count is a confounding variable in model comparison. Second, on a clinically challenging and previously untested task -- distinguishing epilepsy from diagnostic mimickers via interictal EEG -- the diverse checkpoint outperforms the narrow-source checkpoint by +12.3 pp balanced accuracy, the largest gap across all evaluations. Third, systematic inconsistencies between EEG-Bench and EEG-FM-Bench reverse model rankings on identical datasets by up to 24 pp; we identify six concrete sources including split construction, checkpoint selection, segment length, and normalization, showing these factors compound non-additively. |
| title | PRISM: Exploring Heterogeneous Pretrained EEG Foundation Model Transfer to Clinical Differential Diagnosis |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2603.02268 |