Smart phone-based platform for diagnosing asthma and COPD

Introduction: In many parts of the world, pulmonary diseases are underdiagnosed or misdiagnosed because patients lack access to well-trained physicians. Within this context, we have developed a mobile phone-based diagnostic tool to assist in the diagnosis of asthma and chronic obstructive pulmonary disease (COPD). Methods: Using our smartphone-based platform (Chamberlain et al, Proceedings of the NIH-IEEE 2015 Strategic Conference on Point-of-Care Technologies), data were collected from 75 consecutive patients who visited a pulmonary clinic in India and from 12 healthy subjects. Data included recordings of lung sounds at 11 auscultation sites on the body, peak flow meter readings, and questionnaire responses about the presence, duration, and severity of breathlessness, coughing and cigarette smoking history. We used this information to train a logistic regression model to identify patients suffering from obstructive airway disorders. We then trained a logistic regression model to differentiate between asthma and COPD patients. Results: The resulting Area Under the Curve for the Receiver Operating Characteristic curve was 0.97 for distinguishing obstructive airway disorders and 0.95 for distinguishing asthma and COPD. Conclusion: These preliminary results demonstrate that our mobile phone-based tools have good potential for use in screening for pulmonary disease and providing diagnostic support for unskilled health workers in low-resource settings.