Noncontact Monitoring for the Detection of Opioid-Induced Respiratory Depression

Autonomous Healthcare, Inc., Santa Clara CA

Investigators

Abstract

Opioid-induced respiratory depression (OIRD) is a life-threatening event, where its timely detection is critical. Failure to recognize respiratory depression and lack of timely institution of intervention can lead to catastrophic cardiorespiratory arrest, anoxic brain injury, and mortality. Detection of OIRD in patients is as critical in hospitals as is outside hospitals. As a result, monitoring for OIRD is relevant for a wide range of patient populations ranging from surgical patients to those with chronic pain taking high-dose opioid medications at home as well as patients with opioid use disorder (OUD). Our overall goal is to develop a non-contact multi- modal monitoring system for the detection of opioid-induced respiratory depression starting with patients in the hospital general ward. Our proposed technology uses a non-contact set of sensors involving Doppler radar and depth imaging to estimate respiration rate and tidal volume. The system is compact and privacy preserving as processing of the data is performed on the device. The system can also work in the dark and the radar signal can penetrate through clothing and blankets to measure chest wall movements resulting from respiration. Our specific aims are: 1. Development of a Software-Based Monitoring System for Early Detection of Opioid- Induced Respiratory Depression. In this specific aim, we intend to develop a software as a medical device (SaMD) for the detection of OIRD onset using an off-the-shelf depth camera. The effort will involve further development of the algorithm, developing a product-grade software, and the required bench testing in support of regulatory clearance. 2. Validating the Performance of the System for Early Detection of Opioid- Induced Respiratory Depression in a Clinical Setting. Here, our goal is to validate the performance of the developed algorithm as part of a prospective observational clinical study in the post-anesthesia care unit. As a first step, the efficacy of the algorithm in detecting OIRD events has to be established, and hence, a patient population with a high incidence of OIRD is selected to generate the required statistical power without a prohibitively large sample size. Our efforts will culminate in a 510(k) application for a software as a medical device to the FDA. 3. Development of a Software-Hardware Solution with Cloud Connectivity for Early Detection of Opioid-Induced Respiratory Depression. The goal is to develop a software coupled with custom-hardware solution with cloud connectivity and integration capability with the EMR and third-party software. This serves as the prototype of the end-product with capability to interface with a partner’s surveillance software. The real-time and concurrent operation of multiple devices will be tested.