GBBANet, electrocardiographic records database
GBBANet | IC

Ischemic Cardiopathy Database

If you wish to use this database in your research, you must cite these articles:

  1. Ledezma, C., Severeyn, E., Perpinan, G., Altuve, M., & Wong, S. (2014, August). A new on-line electrocardiographic records database and computer routines for data analysis. In Engineering in Medicine and Biology Society (EMBC), 2014 36th Annual International Conference of the IEEE (pp. 2738-2741). IEEE.
  2. Wong, S.,Mora, F., Passariello, G., Silva, J., Almeida, D. (1996) "A new computarized approach for assesing myocardial ischaemia". Revista Brasileira de Engenharia Biomedica. 1996

This work proposed the developement of a stress test analysis method for the diagnostic of the ischemic cardiopathy. For this, a new evaluation criteria and a complete analysis method was created. During this research a database was created containing the data from three ECG channels, acquired in the rehabilitation unit of the Caracas' University Hospital.
The analysis was made on 31 registries of healthy subjects and 42 ischemic patients. The ischemic patients were diagnosed using the traditional criteria by the medical staff, resulting in 19 presenting the sickness and 23 resulting false positives, showing the difficulty in the interpretation of the data.
The processing of the ECG signal included a QRS complex detection, the alignment and averaging of the heartbeats, the mesuring of the ST segment's elevation and slope and the mesuring or extraction of different parameters. These parameters were: ST segment elevation, maximum heart rate, delta ST/HR index, ST segment slope, ST/HR slope. These parameters were hierarchized using discriminant measures and the Fisher discriminant.
It was concluded that the most sensible parameters for the diagnostic were the heart rate, the delta-ST/HR index and the ST/HR slope. Two methods were used for patern recognition: closest mesure and the Fischer clasifier in order to make a preliminary evaluation of the database.
The main result of this work was finding out that the best evaluation method in the stress test is using the heart rate and the delta-ST/HR index as parameters and using a closest mesure clasifier. This method presented a 78.6% sinsibility over the 45.2% obtained using the traditional method, thus demonstraiting the potential of this approach.

In cooperation with the Caracas' University Hospital, ECG registries from derivations II, V5 and V6 were collected under the following conditions:

  1. Two groups of male individuals of ages 53 +/- 9.3 years old. The first was a healthy group (23 subjects) and the other presented coronary desease (27 patients). Both groups performed two stress tests with one week between each test. Patients with recent heart attacks were not included.
  2. There was no medication variation between both tests. No cardiac events or symptom changes happened. Each registry pair was conducted at the same time with the same equipement.
  3. The stress tests were conducted with a QUINTON 3000 or a QUINTON 5000 equipment, using the Bruce protocol without modifications.
  4. The tests were stopped in case of angina pectoris, ST level elevation over 3mm, fatigue or hypertension
For the data acquisition a prototype developped in the GBBA called SISPAS (Signal acquisition system) was used. The main characteristics of this system are:
  1. 3 channels of 12 bit resolution.
  2. Variable sampling frequency: 250Hz, 360Hz, 500Hz and 1000Hz.
  3. Serial RS-232 @ 57600 BPS communication.
  4. Gain control (1, 2, 4, 8) to adjust the siganls to the A/D converter and thus increase the S/N ratio.
The data acquired by the QUINTON was connected to the SISPAS where it was preamplified, filtered, digitalized and transmitted to the Hard Drive Disk. The sampling frequency used in this work was 250 Hz, chosen because it was a standard in databases and due to storage limitations.

Registry reading

This MATLAB function is used to read the registries acquired using the Signal Acquisition System (SISPAS). The function definition is as follows:

[D2 v5 v6 t] = read_reg(name)

The input value of this function is a string containing the full (relative) path to the registry one wishes to read. The string must also contain the .DAT extension at the end of the file, i.e.: name = 'ControlGroup/D151.DAT'.
The return values of this function are the three complete raw signals corresponding to the three derivations plus a time axis that may be used to plot the three signals with a time reference. In addition to this, the function returns a plot showing a part of the three signals that were read. An example of the return plot of the function is:

Plot example for read_reg()

These plots are used to check that the signals were correctly read.
It's important to take into account there has been no preprocessing (other than that made by SISPAS), so the signals may need some processing before they are useful.

Make sure you have read the prior sections before using these signals. Their correct usage and results interpretation may depend on it.
To download the Ischemic Cardiopathy Database signals click here.



Website designed, programmed and maintained by Carlos A. Ledezma

Creative Commons License
GBBANet by Grupo de BioingenierĂ­a y BiofĂ­sica Aplicada - USB is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.