CODEWITHSUNDEEP
python-3.xsignal-processing
Divyayan Dey
Divyayan Dey

Reputation: 15

ECG signal QRS detection with wfdb

I am using the wfdb 3.1.1 for QRS detection in ECG signals. My code goes like this -

record,fields=wfdb.io.rdsamp(r'D:\Research\Biomedical Image Proc\\ECG-Apnea records\a01',sampfrom=200,sampto=700) 
wfdb.plot_items(signal=record,title='ECG Apnea data for patient 1',return_fig=True)   
xqrs=processing.xqrs_detect(sig=record[:,0],fs=fields['fs'])     
print(xqrs)

However, after I plot the detected peaks on the ECG signal, I notice that some of the T peaks are detected instead of the R peak. Is this a recognized issue with the wfdb QRS detection class? The reason I need this is because I need to slice the beats after this. Without proper peaks, the results won't be promising.

Upvotes: 0

Views: 1065

Answers (1)

Léonard
Léonard

Reputation: 2630

Please, note that detecting QRS in ECG data is non trivial task, and still many research are conducted to improve the accuracy of detection algorithms. The questions are:

  • How many QRS detection failure compared to good detection do you obtain (or approximate percentage of wrong detections), and how does this value compare with usual QRS detection algorithms?

  • did you try to perform the detection with different parameters? If no, I recommend visiting the online documentation where you can find plenty of case-dependent fine-tuning settings (in the Conf class for XQRS configuration) like QRS width, max and min values, etc.

Upvotes: 1

Related Questions