Currently, diagnosis of Parkinson's disease doesn't occur until it has progressed to stages that are difficult to treat. Researchers at China's Zhejiang University have developed an artificially intelligent olfactory system, the "e-nose", in hopes of detecting the disease earlier through smelling the skin. Parkinson's Disease patients secrete more sebum from their glands causing an odor. The team tested the device by collecting and analyzing sebum from the backs of Parkinson's patients and healthy control subjects. They found the diagnostic model they used had a 70.8% accuracy in predicting the disease, it rose to 79.2% with the use of machine learning algorithms.