In the JournalsPerspective

Smartphone recordings of coughs accurately diagnose respiratory conditions

Photo of Paul Porter
Paul Porter

A smartphone app that records coughs can diagnose children with respiratory conditions, including asthma, pneumonia and bronchiolitis, with an accuracy comparable to a clinician, according to the results of a prospective, multicenter study.

Paul Porter, MBBS, FRACP , adjunct associate professor in the faculty of health science at Curtin University in Australia, told Infectious Diseases in Children that physicians face numerous challenges diagnosing children with symptoms of respiratory disease.

“Wheezy conditions can include asthma, bronchiolitis, pneumonitis and atypical or viral pneumonia,” he said. “To work these out requires expertise, experience and differential treatment.”

To develop the app, Porter and colleagues recorded cough sounds from 585 children aged 29 days to 12 years in typical clinical environments using a smartphone’s microphone. They developed a diagnostic algorithm that incorporated the cough sounds with five patient- or parent-reported symptoms, including the presence or absence of fever, rhinorrhea, audible wheeze (not stridor), hoarse voice and maximum days of symptoms.

“The objective was to compare diagnoses made by the algorithm to those from a clinical adjudication panel (who had access to all medical records and diagnostic support service results) in order to determine positive and negative percent agreement for a number of respiratory conditions,” the researchers wrote.

Porter said that for 30% of participants whose coughs were recorded, the expert clinicians on the consensus panel — who were “very experienced consultant pediatricians” — disagreed with each other over the diagnosis, showing just how difficult respiratory disease is to diagnose, even with a thorough exam, detailed history and the results of therapeutic testing.

According to the study, the positive percent agreement and negative percent agreement values between the cough analyzer and the clinical reference were 97% and 91% for asthma, 87% and 85% for pneumonia, 83% and 82% for lower respiratory tract disease, 85% and 82% for croup and 84% and 81% for bronchiolitis.

Porter said the technology is ready to use, but he is hoping that it will become more widely accepted within the medical community.

“We are very keen for this tool to be thoroughly accepted and regarded as a serious medical device, not ‘an app on the app stores,’” he said.

Porter and colleagues applied for FDA approval earlier this year. – by Katherine Bortz

Disclosures: Porter is a scientific advisor and shareholder in ResApp Health, who developed the software for the technology used in the study. Please see the study for all other authors’ relevant financial disclosures.

Photo of Paul Porter
Paul Porter

A smartphone app that records coughs can diagnose children with respiratory conditions, including asthma, pneumonia and bronchiolitis, with an accuracy comparable to a clinician, according to the results of a prospective, multicenter study.

Paul Porter, MBBS, FRACP , adjunct associate professor in the faculty of health science at Curtin University in Australia, told Infectious Diseases in Children that physicians face numerous challenges diagnosing children with symptoms of respiratory disease.

“Wheezy conditions can include asthma, bronchiolitis, pneumonitis and atypical or viral pneumonia,” he said. “To work these out requires expertise, experience and differential treatment.”

To develop the app, Porter and colleagues recorded cough sounds from 585 children aged 29 days to 12 years in typical clinical environments using a smartphone’s microphone. They developed a diagnostic algorithm that incorporated the cough sounds with five patient- or parent-reported symptoms, including the presence or absence of fever, rhinorrhea, audible wheeze (not stridor), hoarse voice and maximum days of symptoms.

“The objective was to compare diagnoses made by the algorithm to those from a clinical adjudication panel (who had access to all medical records and diagnostic support service results) in order to determine positive and negative percent agreement for a number of respiratory conditions,” the researchers wrote.

Porter said that for 30% of participants whose coughs were recorded, the expert clinicians on the consensus panel — who were “very experienced consultant pediatricians” — disagreed with each other over the diagnosis, showing just how difficult respiratory disease is to diagnose, even with a thorough exam, detailed history and the results of therapeutic testing.

According to the study, the positive percent agreement and negative percent agreement values between the cough analyzer and the clinical reference were 97% and 91% for asthma, 87% and 85% for pneumonia, 83% and 82% for lower respiratory tract disease, 85% and 82% for croup and 84% and 81% for bronchiolitis.

Porter said the technology is ready to use, but he is hoping that it will become more widely accepted within the medical community.

“We are very keen for this tool to be thoroughly accepted and regarded as a serious medical device, not ‘an app on the app stores,’” he said.

Porter and colleagues applied for FDA approval earlier this year. – by Katherine Bortz

Disclosures: Porter is a scientific advisor and shareholder in ResApp Health, who developed the software for the technology used in the study. Please see the study for all other authors’ relevant financial disclosures.

    Perspective
    Richard F. Jacobs

    Richard F. Jacobs

    The study by Porter and colleagues is a well-designed and controlled study. The positive and negative predictive values are impressive compared with a traditional clinical approach. The use of artificial intelligence and a mobile device adds to the potential promise of technology to help clinicians care for children with respiratory illnesses.

    The use of the app as an adjunct to a physician’s clinical judgment and selective testing (eg, X-ray) could have a very positive impact on the overuse of antibiotics and even decisions for potential treatment. I look forward to additional research. As always, confirmatory studies regarding its use, utility and cost will also be important.

    • Richard F. Jacobs, MD
    • Infectious Diseases in Children Chief Medical Editor
      Professor emeritus
      University of Arkansas for Medical Sciences

    Disclosures: Jacobs reports no relevant financial disclosures.