# First automated detection of a cardiac arrest using a commercially available smartwatch: a case report

## TL;DR

This case report shows for the first time that a smartwatch can detect a real cardiac arrest, using sensor data in a palliative care setting.

## Contribution

The paper presents the first automated detection of a spontaneous cardiac arrest using a consumer-grade smartwatch in a real clinical scenario.

## Key findings

- A commercial smartwatch accurately detected the moment of cardiac arrest in a palliative care patient.
- The detection was validated against clinical reference signals like ECG and PPG.
- This demonstrates the feasibility of using smartwatch data for cardiac arrest detection in real-world settings.

## Abstract

Automated cardiac arrest detection aims to shorten the time between arrest onset and emergency medical services activation, thereby reducing the number of unwitnessed out-of-hospital cardiac arrests (OHCA) and shortening time to treatment in witnessed OHCA. Current arrest detection algorithms are largely developed using simulated or artificially induced cardiac arrest data. To our knowledge, this case report provides the first detailed description of the automated detection of spontaneous, non-procedural, end-of-life cardiac arrest using consumer-grade smartwatch-derived sensor data.

An 82-year-old patient presented to the emergency department with a severe intracerebral hemorrhage with poor prognosis. Following shared decision-making with the family, palliative management was initiated. The patient was continuously monitored with electrocardiography (ECG), invasive arterial blood pressure, and clinical photoplethysmography (PPG). In addition, a commercial smartwatch was placed on the wrist to collect sensor data during the palliative phase and up to 20 min after confirmed cardiac arrest. The smartwatch PPG data were retrospectively analyzed using a previously described diagnostic algorithm. This preliminary algorithm detects circulatory arrest using the photoplethysmography sensor signals acquired from a commercial smartwatch. The algorithm accurately identified the moment of cardiac arrest in concordance with the clinical reference signals. Informed consent was obtained for this research from a legal representative.

Although this controlled end-of-life setting does not represent the circumstances of an OHCA, this case demonstrates the feasibility of detecting true cardiac arrest using a commercial available smartwatch. Prospective studies in real-world OHCA populations are needed to assess clinical performance and practical applicability.