A Non-invasive, Wearable, Miniaturized Auscultation Device for Diagnosis of Pulmonary Diseases

点击翻译按钮获取中文摘要

Summary: Respiratory diseases (RDs) are the fifth cause of death in the US and impose over $150B on healthcare cost. RDs like asthma are incurable and can be life-threatening if not treated promptly. As most dangerous RDs are progressive and manageable with preventive treatment, early detection is crucial to prevent exacerbation. Current diagnosis relies on costly and time-consuming clinical visits, discouraging preventative screening without severe symptoms especially for low/middle-income population at higher RD risks due to more exposure to tobacco and work-related particulates. Diagnosis of RDs like asthma also relies on detection of intermittent abnormal respiratory sounds, which can benefit from prolonged recordings outside the clinic setting. In this project, a low-cost, low-profile, easy-to-use and effective device will be developed to detect adventitial asthma respiratory sounds, facilitating access to screening and continuous treatment monitoring in RD patients. Auscultation is a powerful, non-invasive and well-established method to evaluate health of cardiopulmonary system through sounds of lung and heart. Stethoscopes have been used by physicians for over two centuries in clinics, but they are unsuitable for continuous cardiopulmonary activities monitoring due to large form-factor and dependence on listening skills and experience, prohibiting critical applications like ambulatory monitoring and early detection of RDs in small children. A need exists for a low-profile, miniaturized, high-precision diagnostic device that is more accessible and can accurately detect and quantify respiratory abnormalities over prolonged measurements without relying on the skills and experience of a physician to interpret the sounds. To that end, the electronic interface and acoustic coupling of a MEMS-based accelerometer contact microphone (ACM) onto skin will be optimized to record respiratory sounds with high fidelity, and compared against clinical diagnosis. Breakthrough, hermetically-sealed, high-precision ACMs with unidirectional vibration sensitivity will be used to overcome limits of standard stethoscopes that are bulky, susceptible to airborne and rubbing noise, and hard to use. Besides lung and heart sounds, the ACM simultaneously acquires respiratory rate, heart rate and physical activities of the users. Data will be analyzed using simple algorithms like time-frequency analysis and continuous wavelet transformation to provide reliable information for diagnosis of asthma by detecting signature sounds like wheezing in a wide frequency range of 100Hz-5kHz. Diagnosis accuracy and comprehensiveness are expected to be improved by the ACM-enabled prolonged recording, capability of correlating respiratory sounds with heart sounds and body motions, and detection of higher frequency signals. Interface between ACM and skin will be optimized to increase acoustic coupling over a wide frequency range for wideband adventitious sounds. Low- profile wearable ACMs will be used in a clinical setting to detect adventitial respiratory sounds indicative of asthma and compared with medical-grade digital stethoscopes and clinician judgment. Potentials of the ACM for detecting information for other RDs like COPD and pneumonia will also be assessed for future developments.