Bridging the Compression-Precision Paradox: A Hybrid Architecture for Clinical EEG Report Generation with Guaranteed Measurement Accuracy

TL;DR

This paper introduces a hybrid system for clinical EEG report generation that ensures measurement accuracy despite extreme data compression, combining signal processing and language models to improve reliability and speed.

Contribution

It presents a novel architecture that separates measurement extraction from text generation, guaranteeing clinical measurement accuracy in automated EEG reports.

Findings

60% fewer false alarms in detection

50% faster seizure detection

Achieved sub-clinical measurement precision

Abstract

Automated EEG monitoring requires clinician-level precision for seizure detection and reporting. Clinical EEG recordings exceed LLM context windows, requiring extreme compression (400:1+ ratios) that destroys fine-grained temporal precision. A 0.5 Hz error distinguishes absence epilepsy from Lennox-Gastaut syndrome. LLMs lack inherent time-series comprehension and rely on statistical associations from compressed representations. This dual limitation causes systems to hallucinate clinically incorrect measurement values. We separate measurement extraction from text generation. Our hybrid architecture computes exact clinical values via signal processing before compression, employs a cross-modal bridge for EEG-to-language translation, and uses parameter-efficient fine-tuning with constrained decoding around frozen slots. Multirate sampling maintains long-range context while preserving event-level precision. Evaluation on TUH and CHB-MIT datasets achieves 60% fewer false alarms, 50% faster detection, and sub-clinical measurement precision. This is the first system guaranteeing clinical measurement accuracy in automated EEG reports.