Fraunhofer IIS detects diseases through body odor

A project at Fraunhofer Institute for Integrated Circuits IIS and Dresden University of Technology has investigated an optical method of distinguishing between patients with various medical conditions based on the volatile organic compounds (VOCs) collected as odor on cotton swabs.

Described in Scientific Reports, the findings could point to novel non-invasive screening technologies for clinical use.

The work builds on detection of VOCs by various means, such as the analysis of exhaled breath. Measurement of body fat metabolism, diagnosis of Heliobacter infection and the detection of cancer markers have been among the goals of optical breath analysis platforms. 

In the case of cancer, specific volatile organic compounds produced as byproducts of the cellular changes underway in malignant tumors can in theory also be found in bodily fluids such as urine and sweat, but their presence in breath offers the easiest route to detecting the fingerprints of the disease.

The Fraunhofer IIS project targeted not urine or sweat specifically, but rather the small molecules emitted through metabolic or pathological processes and generally classed as body odor.

"Diseases can alter the composition of VOCs and thus body odor," commented the project. "However, existing analytical methods are often complex, time-consuming, and require specialized personnel. Consequently, there is currently a lack of rapid, simple methods that can be used on-site for widespread application."

Laser-based photoacoustic spectroscopy (LPAS) is one method known to detect VOCs, based on the conversion of optical absorption into acoustic emission by molecular species of interest and assisted by development of tunable quantum cascade lasers. However, clinical applications of LPAS remain in the experimental stage, noted the team in its paper.

"Many systems are optimized for breath analysis, while applications targeting odors from swabbed materials such as cotton swabs are still underexplored. The body emits certain odors - from skin sebum or sweat - that serve as early diagnostic warning signs. For this reason, swabs are a promising alternative to gas sampling bags and also easier to handle."

Miniaturized diagnostic platforms for clinics

Fraunhofer IIS envisaged using LPAS to test not for specific disease-related VOC biomarkers but rather to examine statistical patterns in the spectral data, patterns that correlate with differentiation between healthy individuals and patients with particular medical conditions.

In proof of concept trials, the project analysed swabs from the nose, navel, and ear of 24 selected patients, and used the statistical method termed principal component analysis to determine the broadly significant changes in VOC composition from the different body parts and the different subjects.

The results showed preliminary evidence that laser-based PAS of cotton swab headspace enables discrimination
between healthy individuals and patients with COVID-19, Parkinson’s disease and mild cognitive impairment, said the project, with identifiable group-level spectral structure observed for each diagnostic group. 

Larger studies involving a more diverse range of participants under real-world conditions will now be needed, to test if the technology could provide early indications of potential diseases and support decisions regarding further testing.

"Spectral class separation can be achieved using as few as five discrete wavelengths, instead of the full-range spectrum," wrote the team in its paper. "This supports the possibility of manufacturing a miniaturized diagnostic device, which is an advantage over well established highly sensitive and specific methods such as gas chromatography–mass spectrometry."