Beijing, 18 June 2025 – In a surprising breakthrough, researchers in China have developed a low-cost, non-invasive method to detect Parkinson’s disease using nothing more than earwax and artificial intelligence—achieving an impressive 94% accuracy rate.
The study, published in ACS Analytical Chemistry, focused on analyzing volatile organic compounds (VOCs) in earwax. These scent-related molecules are altered by the disease’s progression, particularly due to neurodegeneration and systemic inflammation. Unlike skin samples, earwax is protected from environmental contamination, making it an ideal medium for consistent biomarker analysis.
Researchers swabbed the ear canals of 209 participants, including 108 people diagnosed with Parkinson’s. They identified four distinct VOCs—such as ethylbenzene and pentanal—that were consistently elevated or suppressed in patients with the disease.
Using these data, the team trained an AI-powered olfactory system, which successfully distinguished between samples from those with and without Parkinson’s, achieving 94% classification accuracy.
“This method offers a fast, low-cost way to screen for Parkinson’s early—potentially before symptoms even appear,” said lead researcher Dr. Hao Dong.
While current Parkinson’s diagnoses often rely on costly brain scans or subjective clinical assessments, this earwax-based method could allow for mass early screening, especially in resource-limited settings. However, the team notes that this was a single-center pilot study, and further research is needed across diverse populations and disease stages.
The research was supported by China’s National Natural Sciences Foundation, the Zhejiang Province R&D Program, and the Central Universities Fundamental Research Fund.
If validated, this AI-earwax system could pave the way for earlier intervention and improved patient outcomes in one of the world’s fastest-growing neurological disorders.