AI-Powered Breath Sensor Capable of Diagnosing 17 Diseases

AI-Powered Breath Sensor Capable of Diagnosing 17 Diseases

Could going to the doctor for a checkup soon involve taking a breath test? Thanks to the work of a multinational team of researchers, that could soon be a possibility.

Researchers have developed a breath test that is capable of detecting 17 different diseases from the presence of volatile compounds in the breath. A diagnosis of multiple sclerosis, chronic kidney disease, preeclampsia, Crohn’s disease, and even cancer is possible using the device.

The researchers, led by Hossam Haick, came from 14 clinical departments all around the world, all of whom were keen to have a device that could be quickly and easily be used tool to detect certain diseases, many of which produced similar symptoms. Such a device could greatly speed up diagnosis of medical conditions, ensure treatment could be administered quickly, and save countless hours of physicians’ time.

Thanks to their research, the team was able to identify signatures of diseases from the breath of patients, and then combine relevant sensors into a device that could be used as a disease breathalyzer.

The device they created includes a nanoarray of sensors that are each capable of detecting a different component in the breath. The presence, concentration, or combination of different compounds can then be used to determine whether an individual is suffering from a particular disease.

However, to get to that point, the researchers first needed to determine which markers could be used for each disease. The team analyzed the breath collected from 1,404 test subjects, some of which had been diagnosed with one of 17 different medical conditions. Using mass spectroscopy, the team were able to identify various components of the breath that were associated with each disease. Based on their research, the team discovered that each of the 17 diseases produced a different chemical ‘breathprint’ comprising different combinations of 13 different components.

A nanosensor array was then developed which was capable of detecting the presence of those compounds, and an artificial intelligence-based system is then used to determine whether an individual likely has a particular disease or medical condition.

The samples were assessed along with those taken from control groups to determine the presence of different compounds in the breath. The nanosensor array detected the compounds and the artificial intelligence-based system analyzed the results and was able to determine which patients were suffering from a particular disease and which were healthy with 86% accuracy. The device is also capable of detecting the unique breathprints of more than one disease in the same individual.

The study was recently published in the Journal ACS Nano.

Leave a Reply