An Automated Classifier of Harmful Brain Activities for Clinical Usage Based on a Vision-Inspired Pre-trained Framework

2025-07-09

AI Summary

To address substantial inter-rater variability among clinicians, poor generalizability of AI models, and challenges in clinical deployment for identifying pathological EEG patterns, this paper proposes VIPEEGNet—a vision-inspired lightweight pretraining framework for EEG analysis. With only 2.8% of the parameters of state-of-the-art models, VIPEEGNet achieves strong cross-site generalization: AUROC = 0.972, sensitivity = 36.8%–88.2%, precision = 55.6%–80.4%, and ranks among the top two globally on the Kullback–Leibler divergence (KLD) metric. Its key innovation lies in adapting vision-based self-supervised pretraining paradigms to time-series EEG modeling via large-scale supervised training. For the first time, it attains human-expert-level multi-class performance across diverse pathological patterns—including epileptic seizures and focal/diffuse abnormalities—thereby significantly improving diagnostic consistency and clinical accessibility.

Technology Category

• Computer Vision: Multi-modal Vision
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Timely identification of harmful brain activities via electroencephalography (EEG) is critical for brain disease diagnosis and treatment, which remains limited application due to inter-rater variability, resource constraints, and poor generalizability of existing artificial intelligence (AI) models. In this study, a convolutional neural network model, VIPEEGNet, was developed and validated using EEGs recorded from Massachusetts General Hospital/Harvard Medical School. The VIPEEGNet was developed and validated using two independent datasets, collected between 2006 and 2020. The development cohort included EEG recordings from 1950 patients, with 106,800 EEG segments annotated by at least one experts (ranging from 1 to 28). The online testing cohort consisted of EEG segments from a subset of an additional 1,532 patients, each annotated by at least 10 experts. For the development cohort (n=1950), the VIPEEGNet achieved high accuracy, with an AUROC for binary classification of seizure, LPD, GPD, LRDA, GRDA, and "other" categories at 0.972 (95% CI, 0.957-0.988), 0.962 (95% CI, 0.954-0.970), 0.972 (95% CI, 0.960-0.984), 0.938 (95% CI, 0.917-0.959), 0.949 (95% CI, 0.941-0.957), and 0.930 (95% CI, 0.926-0.935). For multi classification, the sensitivity of VIPEEGNET for the six categories ranges from 36.8% to 88.2% and the precision ranges from 55.6% to 80.4%, and performance similar to human experts. Notably, the external validation showed Kullback-Leibler Divergence (KLD)of 0.223 and 0.273, ranking top 2 among the existing 2,767 competing algorithms, while we only used 2.8% of the parameters of the first-ranked algorithm.

Problem

• Automated detection of harmful brain activities using EEG
• Overcoming limitations in AI model generalizability and variability
• Validating high-accuracy classification for multiple brain activity categories

Innovation

• Vision-inspired pre-trained CNN model VIPEEGNet
• Validated on large multi-expert EEG datasets
• Achieves expert-level accuracy with minimal parameters