R package for ECGDT (ECG Diagnosis Tool). A system that is capable of:
- Load digitized ECG records.
- Show the ECG signal in the different channels available.
- Detect beats, using single-channel and multi-channel methods.
- Show the ECG signal, with the detected beats highlighted.
- Create a basal (average) beat for a given one.
- Delineate waves inside a beat.
- Obtain a diagnosis score for six major diseases (bradycardia, ischemia, myocardial infarction, tachycardia, ventricular hypertrophy and Wolff-Parkinson-White syndrome), indicating the possibility of their presence in the current record.
(*) ECGDT directory contains the GUI files, and the RDS objects needed with the diagnosis models for each disease.
(*) R directory contains R code files, with the core functions of the tool.
(*) src directory contains C code files, with the more efficient code for more complex tasks.
If you want to run an ECGDT test as in the demo video (https://youtu.be/wh4JW_BOmJk), you can download a 12-lead ECG record from PhysioNet (https://physionet.org/content/incartdb/1.0.0/). For example, if you use record I17:
-
First, download I17.atr, I17.dat and I17.hea files.
-
Then, you can obtain an ASCII file using the WFDB applications, running the command:
rdsamp -r I17 > I17. -
After that, you can add to the resultant ASCII file the header in ECGDT format, by pasting at the beginning of the file:
<Header>
id=I17
sFreq=257
nLeads=12
nSamples=462600
channels=I,II,III,AVR,AVL,AVF,V1,V2,V3,V4,V5,V6
gender=M
conditions=bradycardia,bundleBranchBlock,ischemia,PVCs
</Header> -
With your record file ready, you can select it from the Load ECG menu.
-
After that, follow the steps to analyze the ECG record.