Texas semiconductor company Ambiq has introduced HeartKit, an open-source AI model using multi-head neural networks purpose-built to enable real-time heart-monitoring applications.

Part of its NeuralSpot Model Zoo, this optimised AI model enables the running of various real-time heart-monitoring applications to help users and their healthcare providers quickly identify any irregular events. As with all Ambiq Model Zoo components, HeartKit includes scripts and tools to help AI developers add real-time ECG monitoring capabilities to their health-tech applications.

Personalised health monitoring is becoming ubiquitous with the development of AI models, spanning clinical-grade remote patient monitoring to commercial-grade health and fitness applications. Most leading consumer products offer similar ECGs for common types of heart arrhythmia.

HeartKit is a reference AI model that demonstrates analysing one-lead ECG data to enable various heart applications, such as detecting heart arrhythmias and capturing heart rate variability metrics. Furthermore, by analysing individual beats, the model can identify irregular beats, such as premature and ectopic beats originating in the atrium or ventricles.

“Ambiq’s HeartKit may be the most comprehensive open-source TinyML implementation of AI-based heart monitoring for IoT endpoint devices,” said Carlos Morales, vice president at Ambiq. “The highly optimised AI model will help developers enable health-tech applications on the Ambiq Apollo 4 Plus SoC in a matter of minutes.”

By leveraging a modern multi-head network architecture coupled with Ambiq’s low-power SoC, the model is designed to be efficient, explainable and extensible. While the pre-trained model is ready to use on Ambiq platforms, it also includes software to train, convert and deploy customised models where needed.

The HeartKit has been released under the permissive BSD-3 licence for ease of deployment and development. It is available now as a Technical Preview at GitHub to download and start developing AI.