PhysioNet/CinC 2004 Challenge - Event B
For further details please see the following paper in the Proceedings of Computers in Cardiology 2004:
"Predicting the end of an Atrial Fibrillation Episode: The PhysioNet Challenge"
F. Cantini (1), F. Conforti (1), M. Varanini (1), F. Chiarugi (2), G. Vrouchos (3)
(1) CNR, Institute of Clinical Physiology, Pisa, Italy
(2) ICS-FORTH, Heraklion, Crete, Greece
(3) ICU-CCU Departement, Venizeleio-Pananeio Hospital, Heraklion, Crete, Greece
---------------------------------------------------------------------------------------
These programs are free software; you can redistribute them and/or modify them under the terms
of the GNU General Public License as published by the Free Software Foundation; either
version 2 of the License, or (at your option) any later version.
You can find the GNU General Public Licence in the site http://www.gnu.org/copyleft/gpl.html or
in the file "GNU GENERAL PUBLIC LICENSE.pdf" attached to this package.
---------------------------------------------------------------------------------------
Run EventB.m to elaborate the whole aftdb database
Input parameters:
mainPath_in -> Atrial fibrillation test database Main Directory
mainPath -> Main Data Directory
Records list with S or T classification is provided at standard output
EventB.m runs matlab script needed to classify the records as S or T.
---------------------------------------------------------------------------------------
QrsAvgCancelTestClp(mainPath_in, casiDir, Record, mainPath) performs the QRST clustering to
detect the signals from the same patient.
Input parametrs are
mainPath_in -> The Atrial fibrillation test database Main Directory as provided by PhysioNet
casiDir -> The subdirectort of mainPath_in where the records are
Record -> The record name (without extension)
mainPath -> The data output directory
The output is a .mat structure containing the QRST classes
---------------------------------------------------------------------------------------
TSmerge (mainPath, casiDir, record_list) perform QRST classes analysis o join the records
from the same patient.
Input parameters are
mainPath -> The data output directory
casiDir -> The subdirectort of mainPath where the structure containing the QRST classes data is
(ths same name as csaiDir in QrsAvgCancelTestClp)
record_list -> A list (in matlab sense) of records (without extension)
The output is a list of records couples at standard output
---------------------------------------------------------------------------------------
QrsAvgCancelTestClp_SC(mainPath_in, casiDir, Record, mainPath) performs the QRST cancellation
in order to extract atrial activity signal (AAS) from the surface ECG signal.
Input parametrs are
mainPath_in -> The Atrial fibrillation test database Main Directory as provided by PhysioNet
casiDir -> The subdirectort of mainPath_in where the records are
Record -> The record name (without extension)
mainPath -> The data output directory
The output is a .dat file to be used by SpettriEvB.m
---------------------------------------------------------------------------------------
SpettriEvB(Path_in,casiDir,Record_1,Record_2) performs the frequency AAS analysis on the 56 to 58
and 58 to 60 seconds intervals to choose the terminating (T) record between Record_1 and Record_2
Input parametrs are
Path_in -> The data directory where the .dat files from QrsAvgCancelTestClp_SC are
casiDir -> The subdirectort of Path_in where the .dat files from QrsAvgCancelTestClp_SC are
(ths same name as csaiDir in QrsAvgCancelTestClp)
Record_1 -> The record 1 name (without extension)
Record_2 -> The record 2 name (without extension)
The output is a score. If score > 0 Record_1 is T-type and Record_2 is S-type;
If score < 0 Record_1 is S-type and Record_2 is T-type