Electronic nose could sniff out Parkinson’s disease

An artificially intelligent olfactory (AIO) system has the potential to detect Parkinson’s disease (PD) through the odour profile of sebum, as shown in a study.

Designed to work like the human nose, the AIO device combines gas chromatography with a surface acoustic wave sensor, which measures gaseous compounds through their interaction with a sound wave. The device is embedded with machine learning algorithms that function to classify the sebaceous skin gas of PD patients based on the peaks in the chromatogram.

To identify significant features unique to PD, the investigators collected sebum samples from 31 PD patients and 32 healthy controls by swabbing their upper backs with gauze. From the samples, the AIO system identified three potential odour biomarkers for PD. VOC peaks with average retention times of 5.7, 6.0, and 10.6 seconds corresponding to octanal, hexyl acetate, and perillic aldehyde, respectively, were significantly different between the PD and control groups. [ACS Omega 2022;doi:10.1021/acsomega.1c05060]

When the three compounds were incorporated in the PD odour diagnosis model, it achieved an accuracy of 70.8 percent in predicting PD. The corresponding sensitivity was 91.7 percent, but the specificity was low at only 50 percent, indicating a high rate of false positives.

When pattern-recognition algorithms were applied to analyse the entire odour profile, the accuracy of the AIO device in diagnosing PD improved to 79.2 percent.

Sebum gas has been reported to be useful in the detection of PD, given that seborrheic dermatitis is a typical nonmotor symptom of PD. In PD patients, increased sebum secretion promotes the proliferation of Malassezia yeasts, which in turn leads to sebum inflammation, the investigators noted. [Clin Cosmet Invest Dermatol 2017;10;87-92]

“Volatile organic compounds (VOCs) such as perillic aldehyde and eicosane may change with the increase of sebum secretion, and the interaction between sebum and the yeast of the microbiome can make human skin smelly,” they added. [ACS Cent Sci 2019;5:599-606; ACS Cent Sci 2021;7:300-306; Nat Commun 2021;12:1592; medRxiv 2021;doi:10.1101/2021.05.29.21258035]

The promise of early diagnosis

Currently, PD is diagnosed based on medical history and physical examination, with no objective and consistent approach. This poses a problem, according to the investigators, because by the time of diagnosis, the disease may have progressed to the middle and late stages.

“PD is a chronic neurodegenerative disease caused by the loss of dopaminergic neurons in the substantia nigra. Therefore, when PD patients develop motor symptoms, they have already lost dopaminergic neurons,” they pointed out. [Mol Pharmacol 2008;74:1345-1358]

“The neurodegeneration process may be too fast, so it is essential to identify PD before widespread neuron loss occurs, and the AIO provides a possible solution,” the investigators said. “This [electronic nose] method presents a new possibility for the early diagnosis of PD.”

Compared with traditional clinical PD diagnosis methods (gas chromatography–mass spectrometry [MS], liquid chromatography–MS, or paper spray ionization coupled with ion mobility–MS), the AIO system improves the detection speed and reduces the detection cost. Being portable and facilitating rapid and noninvasive diagnosis, the system can also be widely used in hospitals, clinics, and homes as a PD patient screening or family self-health check method.

However, the investigators acknowledged that before AIO can be adopted in clinical practice, the system must undergo testing on many more individuals to improve the accuracy of the models, as well as consider factors such as race.