St.-Petersburg Institute of Cardiological Technics 12-lead Arrhythmia Database

When referencing this material, please include the standard citation for PhysioNet:

Goldberger AL, Amaral LAN, Glass L, Hausdorff JM, Ivanov PCh, Mark RG, Mietus JE, Moody GB, Peng C-K, Stanley HE. PhysioBank, PhysioToolkit, and PhysioNet: Components of a New Research Resource for Complex Physiologic Signals. Circulation 101(23):e215-e220 [Circulation Electronic Pages;]; 2000 (June 13).
sample from Holter recording

This database consists of 75 annotated recordings extracted from 32 Holter records. Each record is 30 minutes long and contains 12 standard leads, each sampled at 257 Hz, with gains varying from 250 to 1100 analog-to-digital converter units per millivolt. Gains for each record are specified in its .hea file. The reference annotation files contain over 175,000 beat annotations in all.

The original records were collected from patients undergoing tests for coronary artery disease (17 men and 15 women, aged 18-80; mean age: 58). None of the patients had pacemakers; most had ventricular ectopic beats. In selecting records to be included in the database, preference was given to subjects with ECGs consistent with ischemia, coronary artery disease, conduction abnormalities, and arrhythmias; observations of those selected included:

Acute MI2
Transient ischemic attack (angina pectoris)5
Prior MI4
Coronary artery disease with hypertension   7 (4 with ECGs consistent with left ventricular hypertrophy)
Sinus node dysfunction1
Supraventricular ectopy18
Atrial fibrillation or SVTA3 (2 with paroxysmal AF)
AV block1
Bundle branch block3

These diagnoses were confirmed by enzyme assays, coronary angiography, electrophysiological study, and pressure monitoring where necessary. For each record, the .hea file includes the patient's age, sex, diagnoses, and a summary of features of the ECG. Each .hea file also includes a patient number (1–32) that identifies the source recording; all records with a given patient number were obtained from the same Holter recording. For convenience, this information is also collected in the files record-descriptions.txt and files-patients-diagnoses.txt.

The annotations were produced by an automatic algorithm and then corrected manually, following the standard PhysioBank beat annotation definitions. The algorithm generally places beat annotations in the middle of the QRS complex (as determined from all 12 leads); the locations have not been manually corrected, however, and there may be occasional misaligned annotations as a result.

This database was contributed by the St. Petersburg Institute of Cardiological Technics (Incart), St. Petersburg, Russia. The database was initially developed by Viktor Tihonenko ( and Alexander Khaustov ( Additional verification was done by Sergey Ivanov and Alexei Rivin (Incart). Please refer questions and corrections to Evgeniy Yakushenko

Version history

September 18, 2015:
The header files have been updated to include the correct checksums and initial values for all signals. The signals and annotations have not been altered.