Purpose:
1. This is the last project of this semester, finally!
2. This project will be involved with:
1) ECG data sets
2) Wavelet transform and scalogram
3) Convolution Neural Network
4) Some basic concepts of image processing
3.Please go to the following link and study “Signal Classification with Wavelet Analysis and Convolutional Neural Networks(CNN)*”
*We will go over the concept of Convolution Neural Networks (CNN) in our next lesson.
4.Practice the examples demonstrated in the studying materials and configure a CNN for ECG classification.
5.Randomly download three ECG data sets from public ECG database, use the trained CNN to classify the downloaded data and report the results.
6.Due date: June 30, 2018.
BACKGROUND REIVEW:
Wavelet Analysis
Using continuous wavelet analysis, spectral features evolve over time can be explore and common time-varying patterns can be identified in two signals and perform time-localized filtering. Using discrete wavelet analysis, we can analyze signals and images at different resolutions to detect changepoints, discontinuities, and other events not readily visible in raw data. Signal statistics can be compared on multiple scales and hidden patterns will be revealed by performing fractal analysis of data.
Scalogram
In signal processing, a scalogram is a visual method of displaying a wavelet transform. There are 3 axes: x representing time, y representing scale, and z representing coefficient value. The z axis is often shown by varying the colour or brightness.
Convolution Neural Network
In machine learning, a convolutional neural network is a class of deep artificial neural networks, most commonly applied to analyzing visual imagery. Convolutional networks are inspired by biological processes in that the connectivity pattern between neurons resembles the organization of the animal visual cortex. Individual cortical neurons respond to stimuli only in a restricted region of the visual field known as the receptive field. The receptive fields of different neurons partially overlap such that they cover the entire visual field. Applications of CNN are developed for image and video recognition, recommender systems and natural language processing.
Convolution Neural Network
In machine learning, a convolutio