Abstract
This study introduces the cerumenogram methodology for cancer diagnosis using volatile organic compounds (VOCs) in human cerumen. Using headspace gas chromatography-mass spectrometry (HS/GC-MS), 158 VOCs were identified, with 27 selected by genetic algorithm-partial least squares (GA-PLS) achieving 100% discrimination between cancer patients and healthy controls.
Key Findings
- 158 volatile organic compounds identified in human cerumen
- 27 VOCs selected as biomarkers through GA-PLS analysis
- 100% discrimination accuracy between groups
- Non-invasive alternative to traditional diagnostic methods
- 10-fold cross-validation with 10 repetitions