# Out-of-hospital cardiac arrest detection by a wearable: the first real-life case

## TL;DR

A wearable wristband successfully detected a real-life cardiac arrest event, potentially enabling faster emergency response.

## Contribution

First real-life case of out-of-hospital cardiac arrest detection using a wearable device with photoplethysmography and accelerometry.

## Key findings

- Wearable wristband detected pulselessness and collapse during a cardiac arrest event in seconds.
- Algorithm correctly identified the cardiac arrest and subsequent collapse using sensor data.
- Patient recovered well after the event, demonstrating the potential of automated detection for timely EMS activation.

## Abstract

•Report of first real-life OHCA detected by wristband-integrated sensors.•A photoplethysmography-model detected pulselessness within seconds during cycling.•Accelerometry-based fall detection model confirmed cardiac arrest-related collapse.•Wearable detection may enable timely EMS activation in unwitnessed OHCA.

Report of first real-life OHCA detected by wristband-integrated sensors.

A photoplethysmography-model detected pulselessness within seconds during cycling.

Accelerometry-based fall detection model confirmed cardiac arrest-related collapse.

Wearable detection may enable timely EMS activation in unwitnessed OHCA.

Automated cardiac arrest detection and alerting using wearable technology is an emerging innovation in resuscitation science. The DETECT program aims to develop a wristband that integrates photoplethysmography and accelerometry to identify out-of-hospital cardiac arrest events and alert the emergency medical services. We report the first case in which the algorithm successfully detected a cardiac arrest that occurred during daily life in an implantable cardioverter-defibrillator (ICD) patient.

A 64-year old male with a secondary prevention ICD participated in the DETECT-3 study. As part of the study, he was wearing the CardioWatch wristband during daily life for a two-month period. While cycling, the patient experienced a ventricular fibrillation cardiac arrest that was successfully terminated by the ICD. The cardiac arrest detection algorithm of the CardioWatch correctly recognized the cardiac arrest event and subsequent collapse based on photoplethysmography and accelerometry data during post-processing. The patient was admitted for rhythm observation and made a good recovery.

This first real-life case shows feasibility of cardiac arrest detection using wrist-derived photoplethysmography and accelerometry. Further studies are warranted to validate the algorithm in additional cardiac arrest cases, and to minimize false positive alerts during daily-life use. Once implemented, automated cardiac arrest detection and alerting has the potential to shorten treatment delays and improve survival.