European ST-T Database
Copyright (C) European Society of Cardiology 1991. All rights reserved.
The European ST-T Database CD-ROM contains nearly three hundred hours' worth of
ambulatory ECG recordings, annotated beat-by-beat (nearly one and one-quarter
million annotations in all). The CD-ROM contains four directories:
edb European ST-T Database
valedb VALE Database
lib DB library for use with Microsoft C (small memory model)
bin MS-DOS executable programs for viewing annotated waveforms, and
for extracting data in text format
The files on the European ST-T Database CD-ROM are in the same format as those
on the MIT-BIH Arrhythmia and Polysomnographic Database CD-ROMs and the MGH/MF
Waveform Database CD-ROMs.
The European ST-T Database
The European ST-T Database is intended to be used for evaluation of algorithms
for analysis of ST and T-wave changes. This database consists of 90 annotated
excerpts of ambulatory ECG recordings from 79 subjects (70 men aged 30 to 84,
and 8 women aged 55 to 71; information is missing for one subject). Myocardial
ischemia was diagnosed or suspected for each subject (see reference 1);
additional selection criteria were established in order to obtain a
representative selection of ECG abnormalities in the database. The database
includes 367 episodes of ST segment change, and 401 episodes of T-wave change,
with durations ranging from 30 seconds to several minutes, and peak
displacements ranging from 100 microvolts to more than one millivolt.
Each record is two hours in duration and contains two signals, each sampled at
250 samples per second with 12-bit resolution over a nominal 20 millivolt input
range. The sample values were rescaled after digitization with reference to
calibration signals in the original analog recordings, in order to obtain a
uniform scale of 200 ADC units per millivolt for all signals. (The calibration
signals are not included in the signal files.) The header files include
information about the leads used, the patient's age, sex, and medications, the
clinical findings, and the recording equipment.
Two cardiologists worked independently to annotate each record beat-by-beat and
for changes in ST segment and T-wave morphology, rhythm, and signal quality.
ST segment and T-wave changes were identified in both leads (using predefined
criteria which were applied uniformily in all cases), and their onsets,
extrema, and ends were annotated. Annotations made by the two cardiologists
were compared, disagreements were resolved by the coordinating group in Pisa
(see reference 3), and the reference annotation files were prepared;
altogether, these files contain 802,866 annotations.
The VALE Database
The VALE (VALidation-Ecg) Database is intended to be used for evaluation of
algorithms for arrhythmia detection as well as ST-T analysis. This database
consists of 35 annotated records, selected from ambulatory ECG recordings of 35
subjects. The database includes 101 episodes of ST segment change, 178
episodes of T-wave change, as well as a variety of ventricular and
supraventricular arrhythmias (see references 4 and 5).
Each record is three hours in duration (except for vale09, which is two hours
and 48 minutes in duration) and contains one signal, sampled at 200 samples per
second with 12-bit resolution over a 25-40 millivolt input range. The sample
values were rescaled after digitization, as for the European ST-T Database.
The records were annotated using a procedure similar to that used for the
European ST-T Database, although the criteria for identifying ST-T changes
differed. In all, the reference annotation files for the VALE Database contain
1. Taddei, A., Benassi, A., Biagini, A., Bongiorni, M.G., Contini, C.,
Distante, G., Landucci, L., Mazzei, M.G., Pisani, P., Roggero, N.,
Varanini, M., Marchesi, C.
ST-T change analysis in ECG ambulatory monitoring: a European
standard for performance evaluation.
Computers in Cardiology 14:63-68 (1987).
2. Marchesi, C.
The European Community concerted action on Ambulatory Monitoring.
Journal of Med. Eng. and Techn. 10:131-134 (1986).
3. Taddei, A., Biagini, A., Distante, G., Marchesi, C., Mazzei, M.G.,
Pisani, P., Roggero, N., Zeelenberg, C.
An annotated database aimed at performance evaluation of algorithms
for ST-T change analysis.
Computers in Cardiology 16:117-120 (1989).
4. Taddei, A., Varanini, M., Macerata, A., Marchesi, C., Contini, C., Biagini,
A., Bongiorni, M.G., Mazzei, M.G., Mazzocca, G.F., and Baratto, M.
An annotated database for the evaluation of algorithms for the analysis of
arrhythmias and ischemic events.
Computers in Cardiology 10:191-194 (1983).
5. Taddei, A., Varanini, M., Macerata, A., Baratto, M., Biagini, A., Bongiorni,
M.G., Contini, C., Mazzei, M.G., Mazzocca, G.F., and Marchesi, C.
Performance evaluation of algorithms for the QRS morphology classification
and for the detection of the ST changes in ambulatory monitoring.
Computers in Cardiology 11:277-280 (1984).
6. Taddei, A., Biagini, A., Distante, G., Emdin, M., Mazzei, M.G., Pisani, P.,
Roggero, N., Varanini, M., Mark, R.G., Moody, G.B., Braaksma, L.,
Zeelenberg, C., Marchesi, C.
The European ST-T Database: development, distribution, and use.
Computers in Cardiology 17:177-180 (1990).
7. Taddei, A., Distante, G., Emdin, M., Pisani, P., Moody, G.B., Zeelenberg, C.,
and Marchesi, C.
The European ST-T Database: standard for evaluating systems for the analysis
of ST-T changes in ambulatory electrocardiography.
European Heart Journal 13: 1164-1172 (1992).
8. ANSI-AAMI EC38-1994. Ambulatory electrocardiographs.
Association for the Advancement of Medical Instrumentation, 3330 Washington
Boulevard, Suite 400, Arlington, VA 22201-4598, USA (1994).
9. Jager F.
Guidelines for assessing performance of ST analysers.
Journal of Medical Engineering & Technology 22:25-30 (1998).
The sample record
This directory contains a twelve-minute excerpt (record x_edb), taken from
the beginning of record e0105 of the European ST-T Database. This excerpt shows
a complete episode of transient ST change in both signals, accompanied by
a T-wave morphology change in signal 0. The annotation file, `x_edb.atr',
contains beat, rhythm, and signal quality labels as in the MIT-BIH Arrhythmia
Database, with additional annotations to document ST and T changes. To print
only the ST and T change annotations, use the command:
rdann -r x_edb -a atr -p s t
The output appears as:
4:00.444 60111 s 0 0 0 (ST0+
4:29.268 67317 t 0 0 0 (T0+
4:55.892 73973 s 0 0 0 (ST1-
8:28.544 127136 t 0 0 0 AT0+250
8:30.928 127732 s 0 0 0 AST0+600
8:34.176 128544 s 0 0 0 AST1-150
9:41.336 145334 t 0 0 0 T0+)
9:58.580 149645 s 0 0 0 ST1-)
10:02.816 150704 s 0 0 0 ST0+)
The last column shows the contents of the `aux' field of each of these
annotations (see the ECG Database Programmer's Guide for general information
about the `aux' field). In this case, the first, fourth, and last of these
annotations indicate the beginning (`(ST0+'), extremum (`AST0+600'), and
end (`ST0+)') of an episode of ST elevation in signal 0. The annotation for
the extremum indicates that the maximum elevation was 600 microvolts (relative
to the baseline for this subject represented by the first 30 seconds of the
record). Similarly, the `(T0+', `AT0+250', and `T0+)' indicate an episode
during which the T-wave amplitude increased by 250 microvolts in signal 0,
and the `(ST1-', `AST1-150', and `ST1-)' indicate an episode of ST depression
in signal 1, with an extreme deviation of -150 microvolts from baseline.
Additional information about this database is available from:
National Research Council (CNR) Institute of Clinical Physiology
Biomedical Engineering and Medical Informatics
56126 PISA, Italy
Telephone: +(39) 050 502771 (501145) Telefax: +(39) 050 503596
George B. Moody
Harvard-MIT Division of Health Sciences and Technology
MIT Room E25-505A
77 Massachusetts Avenue
Cambridge, MA 02139 USA
Telephone: +1(617) 253-7424 Telefax: +1(617) 258-7859