Respiratory failure due to pneumonia is the leading cause of death in infants and children under 5 years of age in low and middle-income countries. These deaths could be prevented if warning signs of respiratory distress were detected early. Clinical parameters of respiratory distress in infants include increased respiratory rate (RR), marked xiphoid retractions (XR), and thoracoabdominal asynchrony (TAA). In practice, only RR is measured routinely, whereas TAA and XR are assessed subjectively from visual observation of infants' breathing. Remarkably, the only clinical tools that exist to evaluate respiratory distress are scoring systems that rely on subjective visual assessment of an infant, so the results are poorly reproducible and dependent on clinicians' training and experience. Our over-arching goal is to develop a mobile health tool to objectively measure respiratory distress in infants. Such a low-cost, accurate, easy-to-use, and readily-available tool could have a significant impact on decreasing child mortality in resource-poor settings, where respiratory diseases continue to be the main cause of death. This study will collect pilot data and validate the feasibility of measuring clinical parameters of respiratory distress (TAA, RR, and XA) from video. The novelty of the proposed study is that it will use the video capture functionality of a commodity smartphone to quantify major signs of respiratory distress objectively, noninvasively, and without the need for any specialized hardware. The accuracy of video-based parameters will be determined by comparing them with the current gold standards. The results of this study will be critical for conducting a larger prospective study on infants in distress that will provide the basis for building a mobile health tool to determine the presence and severity of respiratory distress.
A low-cost, accurate, and readily-available mobile health tool to objectively measure respiratory distress could have a significant impact on decreasing child mortality in resource-poor settings. Increased respiratory rate, marked xiphoid retractions, and thoracoabdominal asynchrony are major signs of respiratory distress. With this study, we will collect pilot data and validate the feasibility of measuring these signs of distress from video recorded by a commodity smartphone.
