Test-bed

A test bed can be setup using C. The data set used for testing in this module was a subset of the MIT-BIH Noise Stress test available online at www.physionet.org. To evaluate sensitivity to noise, one can read different noise level recordings; here, we analyze three in particular from the Physionet Noise Stress Test tool, selecting 3 different Signal to Noise Ratios (SNR): 24 dB, 18 dB, and 0 dB.

First, download all of the compilers and specialized libraries for gathering ECG data, initialize storage spaces, processing signals stored in a certain format, formatting outputs for consistency and re-use in other algorithms, etc. This is all contained in the WFDB library toolkit. Then, download a specific heart rate file of your choice from the Physionet database and run one of the C programs containing one of the available algorithms under the physiological signal processing header. These algorithms are designed to output a Physionet-compatible annotation file. This annotation file can then be processed manually by running the instantaneous heart rate algorithm (ihr). Using this setup also allows one to save the output of the algorithm as a regular annotation file. This ensures that in the future, the setup could be used to read annotation files and save them directly onto a memory storage unit for processing elsewhere.

The test here constructed consists of running each one of the three algorithms on 3 different data files from the noise stress test. The annotations are compared to reference annotations available on the Physionet-website. Using Physionet’s “bxb” program, one can compare the annotations beat by beat. The bxb program outputs the two key parameters of interest in the analysis of heart beat detection: Sensitivity and Positive Predictivity.

Sensitivity (Se) = TP/(TP+MB)

Positive Predictivity (+P) = TP/(TP+FP)

TP: number of true positive detections.

MB: number of false negatives or missed beats

FP: number of false positives; the system detected a beat where the reference annotation showed no beat.

Results

Table 1

As can be seen from Table 1, the WQRS algorithm is overall more robust and tolerant to noise than the SQRS algorithm, particularly in the Sensitivity ratio. To note, however, is the situation that occurs to WQRS at 0dB noise, wherein the SQRS actually obtains a better Positive Predictivity ratio. This indicates that the WQRS formula is particularly adept at correctly noting beats and rarely misses beats, but is less capable of avoiding false positives at high noise levels. However, the WQRS also suppresses information in the ECG signal that we do not need i.e. the P and T waves which also lends to the filters effectiveness. Given that the overall spread from the other noise levels favors WQRS more, it seems best to side with WQRS for most applications.

Content actions

Add module to:

My Favorites Login Required (?)

'My Favorites' is a special kind of lens which you can use to bookmark modules and collections. 'My Favorites' can only be seen by you, and collections saved in 'My Favorites' can remember the last module you were on. You need an account to use 'My Favorites'.

| A lens I own Login Required (?)

Definition of a lens

Lenses

A lens is a custom view of the content in the repository. You can think of it as a fancy kind of list that will let you see content through the eyes of organizations and people you trust.

What is in a lens?

Lens makers point to materials (modules and collections), creating a guide that includes their own comments and descriptive tags about the content.

Who can create a lens?

Any individual member, a community, or a respected organization.

What are tags?

Tags are descriptors added by lens makers to help label content, attaching a vocabulary that is meaningful in the context of the lens.

| External bookmarks

Footer

This work is licensed by Sharon Du and Dan Calderon under a Creative Commons Attribution License (CC-BY 3.0), and is an Open Educational Resource.

Last edited by Collins Williamson on Apr 15, 2011 3:50 pm -0500.