Bp 361 en

From DCEwiki
Jump to navigation Jump to search

Quality assessment for mobile ECG

Autor: Tadeáš Odstrčilík

Bakalářské práce 2012

Download thesis in PDF

Bachelor thesis follows a PhysioNet/Cinc Challenge 2011: Improving the quality of ECGs collected using mobile phones. The task is to develop an algorithm determining quality of 12 lead ECG. I have developed an algorithm determining the quality of the ECG by using simple rules (e.g., zero signal or high-amplitude artifacts). My method performs assessment of the signal using perceptron and k-nearest neighbor algorithms. Next, algorithm using di erent transformations of ECG. It provides possibility of extending a set of features. The combination of weak classifiers is provided by the AdaBoost algorithm. In the conclusion I have developed scoring system to support assessment of ECG. The signal from each lead is separated into bins with fixed length using Scoring system. Applicability of bins determines the level of quality of ECG recording.

Bp 2012 odstrcilik tadeas.pdf