EEG Processing Pipeline for Objective ASD Diagnosis

Authors

Lord Charité Igirimbabazi, Dartmouth College
Ian Kamweru, Dartmouth College
Ly Nguyen, Dartmouth College
Evan Rankin, Dartmouth College
Ben Sheldon, Dartmouth College

Year of Graduation

2025

Sponsor

Neurizons

Project Advisor

Alexander Boys

Instructor

Solomon Diamond

Document Type

Report

Publication Date

2025

Abstract

15-20% of the world's population exhibit some form of neurodivergence. One neurodivergent disorder, Autism spectrum disorder (ASD), affecting 1 in 36 children, relies on subjective behavioral assessments for diagnosis. This is because modern screening tools have high false positive and false negative rates. Accurately diagnosing ASD early leads to significant improvements in cognitive, language, and social-emotional functioning in children. Early diagnosis enables earlier entry into specialized education programs, improving academic outcomes. Socially, early diagnosis leads to greater independence in adulthood, reducing healthcare costs and the need for intensive interventions. Currently, one biomarker has been accepted into the FDA’s Biomarker Qualification Program for ASD diagnosis via EEG. Additionally, many studies have found differences in EEG data between ASD and non-ASD patients. However, due to high heterogeneity in study results, one consistent method has yet to be adopted as an objective ASD EEG diagnostic. Our sponsor is a healthy tech startup committed to creating an objective autism diagnosis process from EEG analysis. This functional healthcare tool will utilize well-researched processing methods to revolutionize neurodivergence diagnosis.

Dartmouth Digital Commons Citation

Igirimbabazi, Lord Charité; Kamweru, Ian; Nguyen, Ly; Rankin, Evan; and Sheldon, Ben, "EEG Processing Pipeline for Objective ASD Diagnosis" (2025). ENGS 89/90 Reports. 69.
https://digitalcommons.dartmouth.edu/engs89_90/69