Featured Photo by cottonbro studio on pexels.

Electrocardiography (ECG) is the method most often used to diagnose cardiovascular diseases. The recent study demonstrates that an AI is capable of automatically diagnosing the abnormalities indicated by an ECG.

In this post we will review and illustrate how AI applies to ECG analysis to outperform traditional ECG analysis.

Contents:

1. Deep Learning
2. ECGAssess
3. New Apple Watch ECG
4. Summary
5. Explore More
6. Embed Socials

Algorithm:

Deep Learning

• Automatic detection of irregular heart rhythms from ECG 
• Detecting Heart Arrhythmias with Deep Learning in Keras with Dense, CNN, and LSTM
• Automatic ECG diagnosis using a deep neural network

Citation: Ribeiro, A.H., Ribeiro, M.H., Paixão, G.M.M. et al. Automatic diagnosis of the 12-lead ECG using a deep neural network.
Nat Commun 11, 1760 (2020). https://doi.org/10.1038/s41467-020-15432-4

• ECG arrhythmia classification using a 2-D convolutional neural network

Since the CNN model can handle 2D image as an input data, ECG signals are transformed into ECG images by plotting each ECG beat during the ECG data pre-processing step. With these obtained ECG images, classification of seven ECG types is performed in CNN classifier step. The 7 classes are:
Atrial Premature Contraction, Normal, Left Bundle Branch Block, Paced Beat, Premature Ventricular Contraction, Right Bundle Branch Block and Ventricular Escape Beat.

ECGAssess

ECGAssess: A Python-Based Toolbox to Assess ECG Lead Signal Quality

The toolbox contains several algorithms that check the signals to determine their quality. During annotation, a signal was considered acceptable if a reliable heart rate could be detected; otherwise, the signal was considered unacceptable. This is a promising way to perform classification of ECG signal quality.

New Apple Watch ECG

• Cardiologists Share Their Thoughts on (And Skepticism Of) the New Apple Watch ECG Feature
• This API can indicate whether your heart rhythm shows signs of atrial fibrillation (AFib) — the most common type of irregular heartbeat and a major risk factor for stroke — and the irregular heart rhythm notification (for all Apple Watches) which will alert you of irregular heart rhythms suggestive of AFib. 
• The Apple Watch can be a useful tool for monitoring your heart health, but it has limitations.
• To properly diagnose when someone is in AFib, doctors use 12-lead ECG machines. They use electrodes placed on different parts of the body to evaluate the heart’s electrical activity in three directions (right to left, up and down, and front to back), which provides a clearer picture of its movement through the heart’s four chambers. 

Summary

• DNN technology can learn from, and interpret, the whole ECG, thus handling multiple abnormalities at once.
• The DNN is trained using the three types of waves (P, QRS, T). 
• The power of the DNN is that by looking at the signal as a whole.
• Even though AI tech (e.g. The Apple Watch) is not a replacement for medical care, it has the potential to transform cardiac diagnostics and relieve healthcare systems from an increasing pressure on demand.

Explore More

ML-Assisted ECG/EKG Anomaly Detection using LSTM Autoencoder

Embed Socials