Continuous intro-operative respiratory auscultation in anesthesia

Abstract

Typically, intro-operative respiratory auscultation for ventilated patients is performed periodically by an anesthesiologist using stethoscopes. Continuous monitoring, recording and computer-assisted signal processing of respiratory sounds can provide critical information in assisting diagnosis and anesthesia management. This is especially useful in situations of airway complications such as asthma, pulmonary edema, bronchial intubation, pheumothorax, spasm, tube malfunctioning, etc. Technology development of a sensor and information processing system for continuous respiratory sound monitoring will enhance diagnosis capability that is currently provided by many measured variables such as airway pressure, flow rate, O/sub 2/ and CO/sub 2/ levels, blood O/sub 2/ saturation, etc. This paper reports a preliminary effort towards developing such a system.The system will use multiple acoustic sensors (accelerometers) to measure respiratory sounds at designated auscultation sites. To attenuate noise artifacts in measured signals, noise reference sensors will be applied; Sensor signals will be scaled, synchronized and amplified by a multi-channel sound adaptor, and then inputted into a computer. The signals are first processed for noise attenuation and signal separation. The clarified signals are then combined with other measured signals for information processing and diagnosis.