Particle.news
Download on the App Store
6 ARTICLES
·
last wk.

AI Analyzes Earwax to Detect Parkinson’s Disease With 94% Accuracy

Using four volatile biomarkers detected in earwax, researchers developed a non-invasive AI tool achieving 94% accuracy that now awaits multicenter validation.

Image
Analysis of ear wax to detect Parkinson's Disease would be a cheap, non-invasive way to detect the condition in its early stages
The AIO system, the researchers say, could be used as a first-line screening tool for early PD detection and could pave the way for early medical intervention, thereby improving patient care. Credit: Neuroscience News

Overview

  • Researchers collected earwax from 209 individuals, including 108 diagnosed with Parkinson’s, and analyzed samples using gas chromatography–mass spectrometry and surface acoustic wave sensors.
  • The study identified ethylbenzene, 4-ethyltoluene, pentanal and 2-pentadecyl-1,3-dioxolane as volatile organic compounds that distinguish Parkinson’s patients’ earwax.
  • A convolutional neural network trained on the VOC profiles achieved 94% accuracy in differentiating Parkinson’s disease from healthy controls.
  • The non-invasive earwax test offers a cost-effective, stable alternative to clinical evaluations and expensive neural imaging by minimizing environmental contamination.
  • Investigators plan multicenter validation across diverse ethnic groups and biochemical analyses to confirm the tool’s generalizability and explore underlying disease mechanisms.