Diagnosing lung problems could soon be as simple as talking into a mobile phone app that uses voice-activated technology. The technique, developed by a pulmonary–critical care fellow at Allegheny General Hospital in Pennsylvania, offers a simpler way to diagnose one of the deadliest diseases in the US, using only the patient’s voice.
Dr Obaid Ashraf, collaborating with researchers at Carnegie Mellon University, has devised the voice-based technology that uses a smart tablet to diagnose chronic obstructive pulmonary disease (COPD), which is the fourth leading cause of death in the US. Globally, COPD, a progressive, life-threatening lung disease that causes breathlessness, caused some 3.17 million deaths in 2015, according to the World Health Organisation (WHO).
In the UAE, COPD prevalence is comparable to most other countries, according to a study published by The American College of Chest Physicians.
Currently, patients have to undergo a pulmonary function test that measures lung function at a hospital or an outpatient pulmonary clinic. Spirometry is the most common of the pulmonary function tests, but the procedure is labour intensive and is more commonly done as an annual test for patients with lung disease.
Being an aerosol-generating procedure, spirometry is also now more cautiously used because of the risk of infection from Covid-19.
Dr Obaid’s alternative procedure, on the other hand, offers a remote, non-invasive way to diagnose COPD using the patient’s voice.
"Voice is an important biomarker of many medical conditions; in particular, diseases of the respiratory system which leave signatures in breath and speech. We hypothesise that voice and breath sounds in human speech are strongly correlated to lung function, and using voice as a biomarker enables non-invasive and accurate detection of lung function," the researchers said in presenting the interim results of their study.
This is not the first time human voice is being used as a biomarker of medical condition, particularly diseases of the respiratory system. However, Dr Obaid pointed out that there is a dearth of medical literature on the use of voice assessment to determine lung function and health.
How it works
Dr Obaid’s team has built an algorithm that could predict the forced expiratory volume (FEV) and forced vital capacity (FVC), as well airflow obstruction classification, by analysing a person’s voice. COPD is diagnosed if the ratio of the FEV, which is the amount of air a patient exhales in the first second of expiration, to the FVC, which is the total amount of air a patient exhales during the test, is abnormally low.
The team has developed a voice-based learning software and loaded it into a tablet to record and analyse the patients’ voices. The technology, however, is still in the trial phase. The team plans to later implement the solution on a mobile app, where patients could simply record their voice and breathing manoeuvres on their phones, and the data can then be quickly processed over the cloud to determine COPD.
The researchers said the technology would be ideal for personalised, frequent monitoring and analysis of an individual’s lung function to enable the development of clinical decision support systems and advance personalised medicine for chronic and acute respiratory care.
"This technology requires no additional custom-built hardware, is cost effective, non-invasive and practical for ubiquitous and frequent use," the researchers said. "Voice and breath data is an ideal medium to detect and monitor lung functioning, especially respiration impairment at scale. The technique offers the promise of widespread, affordable medical screening and real-time monitoring of respiratory disease."