Home Community User Groups Special Interest Groups Biomedical User Group Biomedical User Group Documents Document

Biomedical User Group Documents

cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 

Using this kit i want to identify the normal and abnormal ECG signal.How can i do that??

by Member on
‎05-09-2012 11:08 AM
- edited on ‎01-27-2017 02:33 AM by: Member kmohsin

I have data downloaded from MIT-BIH database.Using the BioMedical Start up kit i can extract the features of the signals.I want to then classify them as normal and abnormal and put the readings under the two classes.How can i do that?

Any suggestions..Thanks

0 Kudos
Comments
ZJ.Gu
Active Participant
Active Participant
on
‎05-10-2012 09:14 AM
‎05-10-2012 09:14 AM

Hi,

In Biomedical Startup Kit, we only provide the functionality to extract the commonly used features from ECG signal.

For classficiation, there are many ways for you to choose. For example, if you are finding those PVCs (premature ventricular contraction), usually those heart beats don't have P wave and have a wider QRS complex (>0.12s).  Thus you could identify PVCs using PR interval and QRS duration.

You could also use clustering or machine learning method to extract several classes from all the heartbeats. Usually the class with largest amount is normal ECG. On NI Labs (ni.com/labs), you could find Machine Learning Toolkit which may help you.

Thanks!

ZJ

0 Kudos
Contributors