In recent years, smell detection of diseases has attracted growing attention in the field of medical research: from dogs (and even bees) trained to sniff out Covid-19 to human “super sniffers” able to detect neurodegenerative diseases in other people. Now, a group of scientists has developed an “electronic nose” that can detect Parkinson’s disease long before the onset of observable symptoms.
How do you detect Parkinson’s disease?
Patients with Parkinson’s typically secrete a higher-than-normal amount of sebum: the oily secretion produced by the skin’s sebaceous glands. This sebum, which is what keeps our skin and hair lubricated, also contains abnormal amounts of volatile organic compounds that give the substance a particular odor.
Apart from a few notable exceptions, that odor cannot be detected by the human nose. With that said, it can be analyzed through a gas chromatography-mass spectrometry process, although doing so requires bulky and costly equipment that is typically only found at designated facilities.
The e-nose
In an effort to enable regular clinics to benefit from such technology, scientists at China’s Zhejiang University have developed a portable prototype that combines gas chromatography technology with a surface acoustic wave sensor. The latter device is typically used to measure gaseous compounds based on how they interact with a sound wave.
As part of the study, the researchers used the e-nose to analyze swab samples from the backs of 31 patients with Parkinson’s disease and 32 healthy volunteers. With the help of machine learning algorithms, the device was able to detect three odor compounds, which had significant differences in concentrations between the two groups.
The team then used the device to analyze 24 more samples, half of which were from patients with Parkinson’s disease, while the other half were from people without Parkinson’s. The device had a 70.8 percent accuracy at detecting which individuals had the disease. The accuracy increased to 79.2 percent when the algorithms analyzed the samples’ entire odor profile.
The scientists are now working on testing the device on a larger number of participants, which they think should increase the e-nose’s accuracy even further.
Source study: ACS Omega — Artificial Intelligent Olfactory System for the Diagnosis of Parkinson’s Disease
