Using Machine Learning to Enhance the EEG Screening Review by Pre-Screening the EEG

Abstract

Artificial intelligence (AI) is increasingly being used to assist physicians with their medical diagnosis and patient care and improve communication between doctors and their patients. This paper demonstrates how machine learning (ML), a subset of AI, can pre-screen electroencephalograms (EEGs) to assess the overall quality of the EEG and identify artifacts or abnormalities. We developed the “Brain Panel,” an automated ML tool that supplements clinical neurophysiologists’ quality review screening. The Brain Panel generates results prior to visual inspection or quantitative EEG (QEEG) analysis, highlighting potential technical issues or clinical concerns through novel metrics and interpretation methods. We subjected 100 Brain Panel reports and 100 corresponding physician reports from the same EEGs to independent AI evaluations using Grok and Claude. Results show that the Brain Panel provides a sensitive, neurologically informed method to estimate artifacts, assess EEG quality, detect drowsiness, and indicate the likelihood of brain abnormalities. By identifying optimal combinations of Brain Panel metrics, we created detection algorithms that achieved sensitivities of 89–95% for clinically relevant findings. This approach demonstrates that using AI to analyze and interpret automated Brain Panel reports produces a system with clear clinical value for detecting technical and clinical issues in EEGs, enhancing neurologist productivity without replacing expert judgment. 

Received: 3 July 2025 | Revised: 25 March 2026 | Accepted: 19 May 2026

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work. 

Data Availability Statement

The data that support the findings of this study are openly available in the Online Data Resource https://www.dropbox.com/scl/fo/9lvvnoui19c16oehliarf/AJumW34SIPIwtjmNkRPumCU?rlkey=sw8lxip2dpnlo57eb5ag0w1c3&e=1&dl=0. 

Author Contribution Statement

Thomas Collura: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration. Agostino Rosace: Methodology, Software, Formal analysis. Robert Turner: Conceptualization, Validation, Investigation, Resources, Supervision. David Ims: Conceptualization, Validation, Investigation, Resources. Bill Brubaker: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review & editing, Visualization, Project administration.