Perth-based Resapp Health, the company that developed an innovative cough-interpreting smartphone app that diagnoses certain respiratory diseases such as pneumonia better than a stethoscope, aims to repeat the success of its West Australian clinical studies in the United States.
Preliminary results from Resapp Health’s first clinical trial in adults conducted at Joondalup Health Campus demonstrated the app was able to diagnose chronic obstructive pulmonary disease (COPD) with 96 percent accuracy, asthma at 92 percent accuracy, and pneumonia at 100 percent accuracy.
In the clinical trial, 322 adults were asked to cough into smartphones and the results were equivalent to the ones of a previous pediatric clinical trial conducted at Joondalup Health Campus and Princess Margaret Hospital in Australia.
According to Tony Keating, ResApp CEO and managing director, two results from the earlier pediatric clinical trials are especially positive.
“The first is that the ResApp technology was able to identify lower respiratory tract disease even when an experienced clinician using a stethoscope was unable to,” he said in a news release.
“The second exciting result was the ability to differentiate the cause of pneumonia — that is whether the pneumonia was caused by a bacterial infection or a viral infection. This is difficult to do using existing techniques and is currently very costly and not available in all but the best hospitals,” Keating said.
ResApp’s technology, originally developed by associate professor Udantha Abeyratne at The University of Queensland with funding assistance by the Bill and Melinda Gates Foundation, is based on the premise that cough and breathing sounds carry vital information on the state of the respiratory tract.
The app was developed using a machine learning approach to develop highly-accurate algorithms which diagnose disease from cough and respiratory sounds. The algorithms make the app “learn” and improve its diagnostic capacity over time.
Machine learning is an artificial intelligence technique that constructs algorithms with the ability to learn from data. Signatures that characterize the respiratory tract are extracted from cough and breathing sounds. The coughing sounds also contain more information than the sounds that a stethoscope is able to pick up.
While the new app appears to be a cheaper, faster, and more accurate option to diagnose lung diseases, it is also superior in telehealth scenarios, where patients in remote locations cannot be examined by a stethoscope.
The company is planning to have the app commercially available by early next year, but at the moment, there are no plans to launch an app store. “Today our focus is providing a better tool for doctors,” Keating said.
“Our primary focus is the U.S., initially by delivering a clinically-accurate diagnostic test to telehealth practitioners and also providing a fast, accurate diagnosis in the emergency department. The next major milestone is completion of a U.S. clinical study to mirror our Australian studies.”
Massachusetts General Hospital was selected as the first clinical study site in the United States.