Abstract More than 64,000 Americans died from drug overdose in 2016 and drug overdose is now the most common cause of death for people under 50 years old in the United States. Furthermore, the number of overdose deaths is increasing with the rise of abuse of powerful synthetic opioids, such as fentanyl. In May of 2017 National Institutes of Health (NIH) and National Institute on Drug Abuse (NIDA) directors Drs. Collins and Volkow outlined how research may help reduce the death toll associated with the current opioid epidemic; one of the current critical needs is the development of new overdose-reversal interventions, including wearable technologies that can detect an (impending) overdose from physiological signals to signal for help, or trigger a coupled automated injection of naloxone. Automated detection of overdose is essential because most opioid overdoses occur when individuals are alone and unobserved by family members or first responders. Opioids cause respiration to slow and become irregular due to mu-opioid receptor mediated suppression of respiratory related regions of the brainstem and spinal cord. Importantly, there are characteristic early changes in breathing pattern that indicate a progression towards significant hypoventilation, but there is currently no easy-to –use method or device to measure these patterns non-invasively. Recently, there has been a renewed interest in respiratory monitoring using tracheal sounds. Tracheal sounds originate from the vibrations of the tracheal wall and surrounding soft tissues caused by gas pressure fluctuations in the trachea. These sounds can be collected from a microphone placed over the trachea and analyzed to determine the real-time respiratory rate and an estimate of respiratory flow and tidal volume. We hypothesize that individual trends in tracheal sounds detected by a machine- learning algorithm will provide an early warning sign of the onset of hypoventilation as a result of opioid overdose in humans. The aims of this proposal are to develop a machine learning algorithm that detects impending hypoventilation due to an opioid overdose and to develop an initial design for a miniature wireless tracheal sound sensor.

Project Narrative RTM Vital Signs, LLC is developing a non-invasive Tracheal Sound Sensor for early detection of impending hypoventilation due to an opioid overdose. The sensor will continuously monitor an individual’s respiratory pattern to detect a significant change from baseline, in which case it will contact a caregiver and/or emergency personnel detailing the location and status of the person experiencing an opioid overdose or initiate a coupled naloxone injection. This technology has the potential to prevent a significant number of deaths as a result of opioid overdose by allowing for the timely detection of hypoventilation and administration of naloxone.