Introduction of the World main ECG database

This article is reproduced from the " Simple Life " blog

Original link: http://blog.163.com/gz_aaa/blog/static/37834532201472095350188/

The following is a personal summary of the younger brother, there is inaccurate place please correct me!

At present, the most important international, authoritative ECG database has four: the United States Massachusetts Institute of Science and Beth Israel Hospital jointly established Mit-bih ECG database; AHA arrhythmia ECG database of the American Heart Society; the CSE ECG database of the European Union and the EU st-t ECG database. In addition to the international widely recognized also has sudden cardiac deathholter DATABASE,PTB Diagnostic ECG DATABASE,PAF prediction challengedatabase, such as ECG database.

I. Introduction of the DatabaseAmerican Mit-bih ECG database

The Mit-bih ECG database in the United States is currently the most widely used database in the world, consisting of many sub-databases, each of which contains a particular type of ECG record, with the most applications being the Mit-bit arrhythmia database and the Mit-bit QT database. Since 1999, with the support of the National Research Resource Center and the National Institutes of Health, they have exposed the database to the Internet, and all the data in the Mit-bit database is free to download and use, and many of the research on ECG in China and abroad is based on the database, The database is used as the source of experimental data and the detection standard of various recognition algorithms.

It includes: Mit-bih arrhythmia database, Mit-bih Noise Stress Test database, Mit-bih atrial fibrillation Database, Mit-bih ECG compr Ession Test Database, Mit-bih long-term database, Mit-bih malignant ventricular arrhythmia database, Mit-bih Normal Sinus Rh Ythm database, Mit-bih ST change database, Mit-bih supraventricular arrhythmia Database

Parameters for some sub-Libraries:

Mit-bih Arrhythmia Database

Diagnostics: Arrhythmia

Sample rate: Hz

Resolution: One bit

Lead: 2

Duration of each piece of data: min

Storage formats: Format 212

Mit-bih ST Change Database

Diagnostics: Recorded During exercise stress tests and which exhibit transient ST depression

Sample rate: Hz

Resolution: Bit

Lead: 2

Duration of data per segment: varying lengths

Storage formats: Format 212

Mit-bih Atrial fibrillation Database

Diagnosis: Atrial fibrillation (mostly paroxysmal)

Sample rate: Hz

Resolution: Bit

Lead: 2

Duration of each piece of data: Ten H

Storage formats: Format 212

Website: http://ecg.mit.edu/

AHA Cardiac arrhythmia ECG database

The AHA arrhythmia ECG database was developed by the American Heart Association (American Heartassociation,aha), funded by the National Institute of Cardiopulmonary and Hematology, which was developed to evaluate the efficacy of ventricular arrhythmia detectors.

Diagnosis:

No ventricular ectopy (records 1001 through 1010)

Isolated unifocal PVCs (Records 2001 through 2010)

Isolated multifocal PVCs (records 3001 through 3010)

Ventricular Bi-and Trigeminy (records 4001 through 4010)

R-on-t PVCs (records 5001 through 5010)

Ventricular couplets (records 6001 through 6010)

Ventricular tachycardia (records 7001 through 7010)

Ventricular flutter/fibrillation (records 8001 through 8010)

Sample rate: Hz

Resolution: Bit

Website: https://www.ecri.org/Products/Pages/AHA_ECG_DVD.aspx

EU CSE database

The European Union CSE (Common standards for electrocardiography, General ECG Standard) ECG database contains 1000 short-term ECG records, using 12 or 15 lead, the main development purpose is to evaluate the performance of ECG Automatic Analyzer.

E-mail: [email protected]

EU st-t database

The European Union's ST-T database is a database developed by the European Cardiology Society (European Society Ofcardiology) to evaluate the performance of St-segment and T-wave detection algorithms.

Diagnostics: Myocardial ischemia is diagnosed or suspected for each subject. Additional selection criteria were established in order to obtain a representative selection of ECG abnormalities in the D Atabase, including baseline ST segment displacement resulting from conditions such as hypertension, ventricular dyskinesia , and effects of medication.

Sample rate: Hz

Resolution: Bit

Lead: 2

Duration of each piece of data: 2 h

Storage formats: Format 212

Website: http://www.escardio.org/Pages/index.aspx

Dynamic ECG database of sudden cardiac death

It is estimated that around the world, 400000 people and millions of children are dead every year, so Physionet held a database on sudden cardiac death to support and stimulate electrophysiological research in this important area.

Diagnosis:

Patients with underlying sinus rhythm (4 with intermittent pacing), 1 who is continuously paced, and 4 with atrial fib Rillation. All patients had a sustained ventricular tachyarrhythmia, and most had an actual cardiac arrest.

Sample rate: Hz

Resolution: Bit

Lead: 2

Duration of each piece of data: min

Storage formats: Format 212

Website: http://physionet.org/physiobank/database/sddb/

PTB ECG Diagnostic Database:

The digital ECG database provided by the National Metrology Agency of Germany is aimed at the research and teaching of algorithmic standards. The data comes from the Department of Cardiology at Benjamin Franklin Medical University in Berlin.

Diagnosis:

Diagnostic class	Number of subjects
Myocardial infarction	148
Cardiomyopathy/heart failure	18
Bundle Branch Block	15
Dysrhythmia	14
Myocardial hypertrophy	7
valvular heart disease	6
Myocarditis	4
Miscellaneous	4
Healthy controls	52

Sample rate: 1000Hz

Resolution: (±16.384 MV)

Lead: 16 (14-channel ECG, 1-channel breathing, 1-channel voltage)

Duration per data: varying lengths (mostly 2 min)

Storage formats: Format 16

Website: http://www.ptb.de/

PAF Predictive Challenge Database

The PAF Prediction Challenge database comes from 2001 for an open race to automatically predict paroxysmal atrial fibrillation/flutter (predicting paroxysmal atrial fibrillation, PAF). The significance of the competition is to stimulate and promote the United States in the exploration of this major clinical issue and foster friendly competition and extensive cooperation environment.

Diagnosis: paroxysmal Atrial fibrillation

Sample rate: 128Hz

Resolution: + Bit

Lead: 2

Duration of each piece of data: 5 Min/30 min

Storage formats: Format 16

Website: http://physionet.org/challenge/2001/

Second, the data download and read

About Downloads

In addition to the AHA Arrhythmia ECG database and the EU CSE database, all of the above databases can be downloaded freely from the Physionet website (http://physionet.org/physiobank/database/). AHA Arrhythmia ECG Database 2 samples can be downloaded on the Physionet website and on its official website (https://www.ecri.org/Products/Pages/AHA_ ecg_dvd.aspx) respectively.

About reading

The above database data is divided into two kinds of save format, namely Wfdb signal files format 212 and wfdb signal files format 16, the specific encoding method can participate in the relevant introduction on the Physionet website (http:// physionet.org/physiotools/wag/signal-5.htm).

Physionet provides a simple and convenient web-based conversion tool Physiobank ATM (Http://physionet.org/cgi-bin/atm/ATM) that can extract data directly from each database, and automatically translates to the required. MAT format or other desired format as required by MATLAB.

Introduction of the World main ECG database