Particle.news

Download on the App Store

AI-Based Earwax Test Pinpoints Parkinson’s Markers With 94% Accuracy

Focusing on protected ear canal secretions offers an affordable alternative to subjective clinical evaluations.

Image
Image
Image

Overview

  • Researchers at Zhejiang University collected earwax samples from 209 participants in China and analyzed volatile organic compounds using gas chromatography–mass spectrometry.
  • Four VOCs—ethylbenzene, 4-ethyltoluene, pentanal and 2-pentadecyl-1,3-dioxolane—were found at significantly different levels in Parkinson’s patients versus healthy controls.
  • An artificial intelligence olfactory model trained on the VOC profiles distinguished Parkinson’s disease samples from non-PD samples with 94% accuracy.
  • The non-invasive earwax screening could serve as a first-line tool for earlier diagnosis and timely intervention before motor symptoms emerge.
  • Researchers plan to conduct multi-center trials across diverse populations and clinical settings to validate and generalize the findings.