Chen, Gao & Forno

Acoustic Waveform Respiratory Examination for Pediatric Inflammatory Airway Diseases

Asthma is a chronic respiratory disease characterized by airway inflammation and hyper-reactivity. It constitutes a major challenge for public health, affecting over 330 million people worldwide and 26 million people in the US, including 8.4% of US children. Asthma costs grew from $6.2 billion in 1990 to $81.9 billion by 2013. COPD, another chronic pulmonary disease, affects ~16million people in the US. For these and other pulmonary diseases such as cystic fibrosis (CF), acute and chronic alterations in lung function are closely related to disease severity and prognosis. Yet, we lack reliable tools to monitor lung function beyond the clinic. Mobile health (m-health) is an emerging field that uses medical sensors and mobile solutions to improve disease management and monitoring. M-health can run on smartphones and other mobile computing devices, and thus has marked advantages in accessibility and connectivity. Today ~80% of people in the US owns a smartphone–a 33% increase compared to 2011. There are thousands of mobile health apps, with hundreds more being introduced every month. However, there are very few solutions for lung function monitoring, and most are inaccurate or require the addition of cumbersome or expensive hardware. Our proposal challenges the status quo by leveraging hardware and computing power already built into regular smartphones. With its synergy of mobile sensors, machine learning techniques, and rigorous clinical evaluation, our approach will improve the detection, monitoring, and management of lung diseases.

The proposed research is highly innovative and represents a marked departure from the status quo, namely limited tools for low-cost, home-based, and daily monitoring of lung disease, particularly for childhood asthma. More generally, this study will open a door for management, long-term monitoring and early detection of chronic diseases using modern mobile devices and advanced machine learning techniques. Relevance to COVID-19pandemic: The pandemic has affected >3.3million people and caused >240,000 deaths to date. While multisystemic, COVID-19 dramatically affects the lungs. Yet, we don’t know if there are early changes in lung function and airway mechanics that could predict disease severity: due to the high R0of SARS-CoV-2, studies using traditional pulmonary function tests(PFTs) are unfeasible, as they require in-person contact, use of shared equipment, and can generate aerosols that can carry the virus. Our smartphone-based solution would address that gap. More importantly, our system would also allow physicians to monitor PFTs remotely in patients with other lung diseases –avoiding the need for in-person testing with shared equipment.

Wei Chen, PhD

Associate Professor of Pediatrics, 

Associate Professor, Biostatistics

Associate Professor, Human Genetics

University of Pittsburgh

Wei Gao, PhD

Associate Professor

Department of Electrical and Computer Engineering 

University of Pittsburgh

Erick Forno, MD MPH

Assistant Professor of Pediatrics, Division of Pulmnary Medicine

University of Pittsburgh

UPMC Children's Hospital of Pittsburgh