Skip to content

Home

HeartKit

Documentation: https://ambiqai.github.io/heartkit

Source Code: https://github.com/AmbiqAI/heartkit

Introducing HeartKit, an AI Development Kit (ADK) that enables developers to easily train and deploy real-time heart-monitoring models onto Ambiq's family of ultra-low power SoCs. The kit provides a variety of datasets, efficient model architectures, and heart-related tasks. In addition, HeartKit provides optimization and deployment routines to generate efficient inference models. Finally, the kit includes a number of pre-trained models and task-level demos to showcase the capabilities.

Key Features:

  • Real-time: Inference is performed in real-time on battery-powered, edge devices.
  • Efficient: Leverage Ambiq's ultra low-power SoCs for extreme energy efficiency.
  • Extensible: Easily add new tasks, models, and datasets to the framework.
  • Open Source: HeartKit is open source and available on GitHub.

Please explore the HeartKit Docs, a comprehensive resource designed to help you understand and utilize all the built-in features and capabilities.

Getting Started

Installation

To get started, first install the local python package heartkit along with its dependencies via pip or Poetry:

$ poetry install .

---> 100%

$ pip install heartkit

---> 100%

Usage

HeartKit can be used as either a CLI-based tool or as a Python package to perform advanced development. In both forms, HeartKit exposes a number of modes and tasks outlined below. In addition, by leveraging highly-customizable configurations, HeartKit can be used to create custom workflows for a given application with minimal coding. Refer to the Quickstart to quickly get up and running in minutes.

Modes

HeartKit provides a number of modes that can be invoked for a given task. These modes can be accessed via the CLI or directly from the task within the Python package.

  • Download: Download specified datasets
  • Train: Train a model for specified task and datasets
  • Evaluate: Evaluate a model for specified task and datasets
  • Export: Export a trained model to TensorFlow Lite and TFLM
  • Demo: Run task-level demo on PC or remotely on Ambiq EVB

Task Factory

HeartKit includes a number of built-in tasks. Each task provides reference routines for training, evaluating, and exporting the model. The routines can be customized by providing a configuration file or by setting the parameters directly in the code. Additional tasks can be easily added to the HeartKit framework by creating a new task class and registering it to the task factory.

  • Denoise: Denoise ECG signal
  • Segmentation: Perform ECG based segmentation (P-Wave, QRS, T-Wave)
  • Rhythm: Heart rhythm classification (AFIB, AFL)
  • Beat: Beat-level classification (NORM, PAC, PVC, NOISE)
  • BYOT: Bring-Your-Own-Task (BYOT) to create custom tasks

Model Factory

HeartKit provides a model factory that allows you to easily create and train customized models. The model factory includes a number of modern networks well suited for efficient, real-time edge applications. Each model architecture exposes a number of high-level parameters that can be used to customize the network for a given application. These parameters can be set as part of the configuration accessible via the CLI and Python package. Check out the Model Factory Guide to learn more about the available network architectures.

Dataset Factory

HeartKit exposes several open-source datasets for training each of the HeartKit tasks via the dataset factory. For certain tasks, we also provide synthetic data provided by PhysioKit to help improve model generalization. Each dataset has a corresponding Python class to aid in downloading and generating data for the given task. Additional datasets can be added to the HeartKit framework by creating a new dataset class and registering it to the dataset factory. Check out the Dataset Factory Guide to learn more about the available datasets along with their corresponding licenses and limitations.

  • Icentia11k: 11-lead ECG data collected from 11,000 subjects captured continously over two weeks.
  • LUDB: 200 ten-second 12-lead ECG records w/ annotated P-wave, QRS, and T-wave boundaries.
  • QTDB: 100+ fifteen-minute two-lead ECG recordings w/ annotated P-wave, QRS, and T-wave boundaries.
  • LSAD: 10-second, 12-lead ECG dataset collected from 45,152 subjects w/ over 100 scp codes.
  • PTB-XL: 10-second, 12-lead ECG dataset collected from 18,885 subjects w/ 72 different diagnostic classes.
  • Synthetic: A synthetic dataset generator provided by PhysioKit.
  • BYOD: Bring-Your-Own-Dataset (BYOD) to add additional datasets.

Model Zoo

A number of pre-trained models are available for each task. These models are trained on a variety of datasets and are optimized for deployment on Ambiq's ultra-low power SoCs. In addition to providing links to download the models, HeartKit provides the corresponding configuration files and performance metrics. The configuration files allow you to easily recreate the models or use them as a starting point for custom solutions. Furthermore, the performance metrics provide insights into the model's accuracy, precision, recall, and F1 score. For a number of the models, we provide experimental and ablation studies to showcase the impact of various design choices. Check out the Model Zoo to learn more about the available models and their corresponding performance metrics. Also explore the Experiments to learn more about the ablation studies and experimental results.