According to the study published in the journal Advanced Science, the scientists could deliver a rapid, non-invasive diagnosis by analyzing these molecules using artificial intelligence (AI). The A-patch device based on this proof-of-concept study is currently undergoing clinical trials.
The group intends to incorporate sensors into the patch in future iterations and read the results via a smartphone.
Tuberculosis, often referred to as “consumption,” is ubiquitous in impoverished countries, accounting for 95 percent of cases. In 2019, about ten million people contracted tuberculosis, and 1.4 million died because of the infection.
Tuberculosis bacteria are predicted to infect approximately one-third of the world’s population. As a result, the World Health Organization (WHO) has designated tuberculosis as a “global health emergency” since 1993. While effective tuberculosis treatment is available, detection remains a challenge, with over 3 million cases missed each year.
Tuberculosis’s early symptoms are non-specific, confounding diagnosis. Also, currently available diagnostic procedures are slow and prohibitively expensive or sophisticated for resource-constrained environments. For instance, a sputum smear ($2.60 to $10.50 each examination) is prohibitively expensive in a region where people survive less than $1 per day. Still, a mycobacterial culture test takes 4–8 weeks and at least three patient visits to confirm the diagnosis and start therapy.
The WHO sees an inexpensive and effective tuberculosis test critical to combating the illness. And it is this requirement that Professor Hossam Haick and his team at the Technion’s Wolfson Department of Chemical Engineering address in their ground-breaking study. The researchers, led by Dr. Rotem Vishinkin, developed a sticker patch to be applied on the patient’s arm. The patch, which had an absorbent pouch, gathered chemicals emitted via the skin. These served as the much-desired diagnostic tool.
Dr. Vishinkin, the project’s scientific leader, noted, “our initial studies, done on a large number of subjects in India and in South Africa showed high effectiveness in diagnosing tuberculosis, with over 90% sensitivity and over 70% specificity. In addition, we showed that tuberculosis can be diagnosed through the compounds released by the skin. Our current challenge is minimizing the size of the sensor array and fitting it into the sticker patch.”
The platform they developed is inexpensive, quick, and simple to use and does not require any specialized employees. The group expects that the same method will diagnose other diseases and disorders in the future, bringing accurate diagnosis to distant places of the world.
Clinical trials were undertaken at South Africa’s University of Cape Town and Groote Schuur Hospital, India’s All-India Institute of Medical Sciences, and Latvia’s University of Latvia and Riga East University Hospital.