chapter.md

Book

Leveraging Data Science for Global Health.

This book explores ways to leverage information technology and machine learning to combat disease and promote health especially in resource-constrained settings. It focuses on digital disease surveillance through the application of machine learning to non-traditional data sources. It also explores the applications of such techniques in real world settings.

Title: ECG Processing in Python

Objectives

This workshop aims to introduce users to core methods used to interpret ECGs.

In the hands-on exercise, you will be asked to implement and evaluate these models on a clinical prediction problem.

No prior programming experience is assumed. Basics of the Python language and Jupyter notebook environment will be covered.

Background

The electrocardiogram (ECG) is an effective non-invasive tool used by physicians to inspect the functionality of hearts.

By placing sets of electrodes over a patient's skin at given locations, each potential difference measured by each electrode pair shows the vector

As a physician looks upon a visual ECG diagram and interprets the underlying workings or irregularities of the heart, so too can algorithms be developed to automatically process these signals and extract information.

Exercises: A set of exercises to help the reader understand the concepts

• qrs detection
• beat classification
  • qrs width
    • That algorithm
    • convolution with wavelet
  • qrs deflection direction
  • t wave deflection direction opposite

Uses and Limitations: It is important that each chapter covers both the various use cases and the limitations of the topic covered by the chapter. Ideally, we would appreciate if the authors could “connect the dots” between the chapter and the other materials in the book.

Although feature engineering and parameter tuning is required, fundamental signal processing techniques offer full transparency and interpretability which is important in the medical setting. In addition, the algorithms are relatively inexpensive to compute, and simple to implement, making it highly applicable to remote monitoring applications.

Even simple microcontrollers have enough memory and compute power to implement the beat-classification and qrs detection algorithms in this chapter. Do not need to be trained by GPUs.

References:

• Please use Harvard reference styling.