← Back to the feed
Biotech & healthResearch findings

EEG-SpikeAgent · LLM-Based Program Synthesis for Automated EEG Spike Detection

Original published 2026.07.06 09:19 KSTAdded to Blackdeck 2026.10.26 22:12 KST

What has changed?

A program synthesis framework using LLMs iteratively generates and refines EEG signal features for spike detection.

Key points

  1. Introduces EEG-SpikeAgent, a program synthesis framework using LLMs to automatically generate EEG features for spike detection.
  2. The system iteratively proposes, evaluates, and refines signal features and decision rules based on performance.
  3. Artifact-aware feature generation outperforms spike-only feature search in balanced accuracy and F1 score.

Connect it to your work

Automates EEG feature engineering in an auditable, code-driven manner for clinical and methodological review.

What remains unproven

This research note is based on an abstract; check the original for test conditions and application scope.

Originals used for this note

Links open the original publisher in a new tab.

  1. arxivEEG-SpikeAgent: Agentic Closed-Loop Program Synthesis for Automated EEG Spike Detection