Indian Journal of Medical Informatics

20-11-2008

Indian Journal of Medical Informatics

Java Based Program for Multi-Parametric ECG Analysis Santosh Bhaskaran M.Sc., Pandit Bhalchandra Vidyasagar Ph.D. Biophysics Laboratory, Department of Physics, University of Pune, Pune - 411007 [email protected], [email protected]

ECG Machines From 320USD CONTEC ECG/EKG Machine Order Online Safe & Easy To Use. Check Us Out!

Abstract: A user-friendly Java-based software has been developed to analyze digitally acquired ECG signals. It reads data from the files each corresponding to different leads, reconstructs the ECG signals, identifies the P, Q, R, S and T wave positions, calculates the time intervals and amplitudes and their mean and standard deviation and also plots their histograms for each of these waves for different leads. It pops the analysis performed onto the window and also writes it to a printable file thus helping for quick assessment of ECG records and indicating possible irregularities related to heart rate variability and the variation in the amplitudes. Keywords: Java Swings, ECG

1. Introduction With the advent of the computational infrastructure the medical informatics has assumed an important role in the medical diagnosis and the treatment. Different types of analytical tools have been developed based on biochemical, MATLAB Wavelet electrical and structural aspects of different organs and the parts of human body. Image analysis carried out using the techniques such as x-rays, CT scan, MRI and the PET has become a routine practice. Many commercial Analysis packages are available for this purpose. Measurement and analysis of biopotentials such as ECG, EEG, EMG, ERG Analyze signals and images using and visual evoked potentials have also been developed and are being used. But many of these techniques are wavelet techniques platform-dependent and also do not consider multiparametric analysis. Current softwares are limited to specific ailments like cardiac ischaemia1, cardiac arrythmias 2, RR interval analysis 5 and QRS detection4. On this with MATLAB background an effort is made in the present study to develop a Java-based multiparametric analysis program, which www.mathworks.com is user-friendly and can help the physician to detect the irregularities in the ECG signal. www.ContecMed.com

ECG: Made in Japan We are the Japanese manufacturer of medical equipments. ECG,Ultrasound, www.fukuda.co.jp

Free TeamCity Server 3.1 Remote code inspection: No broken code in your version control. EVER!

2. Materials and Methods A Java-based software using Java Swings as the programming tool has been developed which can analyze previously acquired ECG signals with data available in digital form. Since it is Java-based, it is platform-independent. The ECG records corresponding to each of the leads are stored in files that can be read in DOS or MS Notepad with the end of the name of each file having name corresponding to each lead which are stored in X-Y format. This software reads data from any of the files (leads) selected and reconstructs the ECG signal. Thus the ECG signal for all the leads can be constructed. The program detects the P, Q, R, S and T peaks, obtains P-P, Q-Q, R-R, S-S and T-T time intervals and the corresponding peak amplitudes. The mean and standard deviation values for time intervals and the amplitudes for each of the peaks are calculated. The program also plots the histograms for each of the above parameters. It pops the analysis performed onto the window and also writes it to a printable file thus helping for quick assessment of ECG records and indicating possible irregularities related to heart rate variability and the variation in the amplitudes.

3. Results and Discussion The flow chart for the program may be provided by the authors to the readers on request. The program reads data from the files (leads) that are obtained using data acquisition, reconstructs the ECG signal and analyzes the signal JetBrains.com/TeamCity for all of the leads. The result of the analysis is popped up onto the window and also written to a printable file creating the report. The analysis gives possible indications of ailments that can be diagnosed using ECG. This is depicted in

http://www.iami.org.in/journal1/java.a…

1/3

20-11-2008


the report. The analysis gives possible indications of ailments that can be diagnosed using ECG. This is depicted in the figures Fig.2a to Fig.2j. Fig.2a shows the basic menu system that is designed. Selection of the file (lead) is shown in Fig.2b which results in the reconstruction of the ECG signal and popping up of the analysis performed onto the window as shown in Fig.2c. Fig.2d to Fig.2h show the plotting of histogram for a particular peak of a particular lead. Fig.2i and Fig.2j show the result of selection of a wrong file. Current softwares are limited to specific ailments like cardiac ischaemia1, cardiac arrythmias 2, RR interval analysis 5 and QRS detection4. A Java-based software has been designed for multiparametric analysis of digitally stored ECG signal records. As the program is Java-based, it is platform-independent. The program detects the P, Q, R, S and T peaks, obtains P-P, QQ, R-R, S-S and T-T time intervals and the corresponding peak amplitudes. The mean and standard deviation values for time intervals and the amplitudes for each of the peaks are calculated. The program also plots the histograms for each of the above parameters. It pops the analysis performed onto the window and also writes it to a printable file thus helping for quick assessment of ECG records and indicating possible irregularities related to heart rate variability and the variation in the amplitudes. Currently the software analyses ECG signals that are already acquired in digital form. It has to be developed further so that it can analyse signals that are acquired online. Also it can be used only on a standalone PC. Further development has to be done so that it can be used in a LAN network. This software can also be further developed for applications in other fields of medicine.

Fig. 1a

Fig. 2c

Fig. 2e

Fig. 2g


Fig. 2b

Fig. 2d

Fig. 2f

Fig. 2h

2/3

20-11-2008


Fig. 2i

Fig. 2j

Acknowledgments I would like to thank my lab-mate Dr. Kaveh for his help in medicine and also Mr. Rahul, Bioinformatics Centre, University of Pune, Pune- 411007, my friend Mr. Anirban and the Java forums on the Sun Microsystems site for their help in debugging my program. One of the authors, Santosh Bhaskaran would like to thank CSIR for its financial support and P.B. Vidyasagar AS-ICTP.Trieste, Italy for the award of the associateship. REFERENCES 1. The HealthRx™ CardioView system for enhanced visualization of cardiac ischaemia ( http://141.156.121.114/ ) 2. Ge D, Srinivasan N, Krishnan SM Cardiac arrhythmia classification using autoregressive modeling BioMedical Engineering OnLine 2002, 1:5 (13 November 2002) 3. Wang H, Azuaje F, Jung B, Black N A markup language for electrocardiogram data acquisition and analysis (ecgML) BMC Medical Informatics and Decision Making 2003, 3:4 ( 7 May 2003) 4. QRS detection and waveform boundary recognition using ecgpuwave (http://www.physionet.org/physiotools/ecgpuwave/ ) 5. RR interval time series modeling: A challenge from PhysioNet and Computers in Cardiology 2002 (http://www.physionet.org/challenge/2002/index.shtml ) | Home | About Us | Members | Links | What's New | Contact Us | Our History | IAHI | Mentors | © Indian Association for Medical Informatics, Webmaster - Webmaster Pow ered by Amlamed.com This page last updated: July 2008


3/3