Study of echocardiogram for myocardial infarction using neural networks

Ultrasonic myocardial echocardiogram is a promising approach to diagnosing myocardial infarction. We digitize echocardiographic images to study ultrasound backscatter for scar tissue detection. Neural network analysis is a successful approach to scar detection. We improve the structure of multi-laye...

Full description

Saved in:
Bibliographic Details
Main Authors: Cheng Yi, Micheli-Tzanakou, E., Shindler, D.M., Kostis, J.B.
Format: Conference Proceeding
Language:English
Subjects:
Backscatter
Biomedical engineering
Hospitals
Myocardium
Neural networks
Neurofeedback
Neurons
Optical imaging
Testing
Ultrasonic imaging
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Ultrasonic myocardial echocardiogram is a promising approach to diagnosing myocardial infarction. We digitize echocardiographic images to study ultrasound backscatter for scar tissue detection. Neural network analysis is a successful approach to scar detection. We improve the structure of multi-layer perception combined with the feedback optimization algorithm ALOPEX to make it more general and we also introduce the noisy training method.
DOI:10.1109/IEMBS.1993.978530