QT ZAMAN ARALIĞININ GAUSS KARIŞIM MODELİ VE YAPAY SİNİR AĞI TABANLI TESPİTİ

Year 2017, Volume: 6 Issue: 2, 752 - 762, 31.07.2017

Mehmet İşcan Cüneyt Yılmaz

https://doi.org/10.28948/ngumuh.342038

Cited By: 1

Abstract

   Günümüzde yarı-parametrik tanımlanan yapay
sinir ağı tabanlı olasılıksal yöntemler biyolojik sinyallerin işlenmesi örüntü
tanımasında aktif olarak kullanılmaktadır. Bu çalışmada, EKG sinyallerinin
önemli bir zaman aralığı olan QT süresinin belirlenmesi ve sınıflandırılması
için yarı-parametrik Gauss karışım modeli tabanlı yapay sinir ağı modeli
gerçeklenmiştir. Bu kapsamda, zamana bağlı değişen kalp ritim sinyallerinin,
eğitimi ve sınıflandırılması olasılıksal metotların gözetimli ve gözetimsiz
eğitimi ile tamamlanmış, ayrıca yeni bir fikir olarak karşılaştırma algoritması
statik yapay sinir ağları için sunulmuştur. Önerilen algoritma ile 105
PHYSIONET QT veritabanı verileri ve 4 gerçek denekten alınmış veriler
işlenmiştir. Gerekli eğitimler tamamlandıktan sonra, sunulan algoritma %97,11
hassasiyet, %94,27 pozitif belirleyicilik ve %4,2 hata oranı ile QRS kompleksi
ve T dalgasını saptayabilmiş, ayrıca 3,1 milisaniye ortalama hata değeri ve 5,62
milisaniye standart sapma değeri ile QT zaman aralığını bulabilmiştir.
Sonuçlara göre, önerilen algoritma değişik EKG sinyalleri için yüksek
performansta sınıflama ve ayrıştırma işlemini gerçekleştirebilmiştir.

Keywords

QT ölçümü, T-dalgası, EKG sinyal sınıflaması, yapay sinir ağları, Gauss karışım modeli

GAUSSIAN MIXTURE MODEL AND NEURAL NETWORK BASED DETERMINATION OF QT DURATION

Abstract

   Nowadays,
probabilistic methods based on semi-parametric neural networks have been used
to signal processing in biological signals with individual characteristics. The
main objective of this study was to develop a semi-parametric neural network
based on Gaussian mixture model to perform the QT measurement and
classification. For this purpose, a comparison algorithm evaluating time-series
cardiac signals was established with training by supervised and unsupervised
learning, and the comparison algorithm was presented in order to static neural
networks. The proposed algorithm has been tested on the data from 4 normal
subjects and 105 additional normal data sets from PHYSIONET QT database. After
the improvement by the proposed algorithm, we observed that the QT-measurements
were done with 3.1 milliseconds of the mean values and 5.62 milliseconds of
standard errors, when QRS complexes and T waves are detected at the rate of
97.11% sensitivity, 94.27% positive predictivity and 4.2% error value,
respectively. The results suggested that the proposed algorithm achieved a
classification and discrimination of various ECG signals at a high performance
level.

Keywords

QT measurement, T wave, ECG signal classification, artificial neural networks, Gaussian mixture model

References

• [1] GOLDBERG, R. J., BENGSTON, J., CHEN, Z., ANDERSON, K.M., LOCATI, E., LEVY, D.,"Duration of the QT Interval and Total and Cardiovascular Mortality in Healthy Persons (The Framingham Heart Study Experience)", The American Journal of Cardiology, 67(1), 55-58,1991.
• [2] FRAZIER O.H., “Ventricular Assist Devices and Total Artificial Hearts—A Historical Perspective”, Cardiology Clinics, 21, 1–13, 2003.
• [3] SORIA-OLIVAS, E., MARTINEZ-SOBER, M., CALPE-MARAVILLA, J., GUERRERO-MARTINEZ, J. F., CHORRO-GASCO, J., ESPI-LOPEZ, J., “Application of Adaptive Signal Processing for Determining the Limits of P and T Waves in an ECG”, IEEE Transactions on Biomedical Engineering, 45(8), 1077-1080, 1998.
• [4] LAGUNA, P., THAKOR, N.V., CAMINAL, P., JANE, R., YOON, H.R., BAYES DE LUNA, A., MARTI, V., GUINDO, J., "New Algorithm for QT Interval Analysis in 24-Hour Holter-ECG: Performance Applications", Medical and Biological Engineering and Computing, 28, 67-73, 1990.
• [5] TIRONI, D.A., SASSI, R., MAINARDI, L.T., "Automated QT Interval Analysis on Diagnostic Electrocardiograms", Computers in Cardiology, 33, 353-356, 2006.
• [6] CLIFFORD, G. D., AZUAJE, F., MCSHARRY, P. E., Advanced Methods and Tools for ECG Data Analysis (1st ed.), Artech House Publishers, Boston, USA, 1989.
• [7] MURRAY, A., MCLAUGHLIN, N.B., BOURKE, J.P., DOIG, J.C., FURNISS, S.S., CAMPBELL, R.W., "Errors in Manual Measurement of QT Intervals", Heart Journal, 71, 386-390, 1994.
• [8] SUGA, H., SAGAWA, K., SHOUKAS, A.A., “Load Independence of the Instantaneous Pressure-Volume Ratio of the Canine Left Ventricle and Effects of Epinephrine and Heart Rate on the Ratio”, Circulation Research, 32(3), 314-322, 1973.
• [9] SCHUARTZ, P.J. and WOLF, S., "QT Interval Prolongation as Predictor of Sudden Death in Patients with Myocardial Infarction", Circulation, 57, 1074-1079, 1978.
• [10] WILLEMS, J. L., ARNAUD, P., VAN BEMMEL, J.H., BOURDILLON, P.J., BROHET, C., DALLA VOLTA, S., ANDERSEN, J.D., DEGANI, R., DENIS, B., DEMEESTER, M., "Assessment of the Performance of Electrocardiographic Computer Programs with the Use of a Reference Data Base", Circulation, 71(3), 523-534., 1985.
• [11] LAGUNA, P., JANE, R., CAMINAL, P., “Automatic Detection of Wave Boundaries in Multilead ECG Signals: Validation with the CSE Database”, Computers and Biomedical Research, 27(1), 45-60, 1994.
• [12] ZHANG, Q., MANRIQUEZ, A.I., MEDIGUE, C., PAPELIER, Y., SORINE, M., “An Algorithm for Robust and Efficient Location of T-wave Ends in Electrocardiograms”, IEEE Transactions on Biomedical Engineering, 53(12), 2544-2552, 2006.
• [13] GOLDENBERG, I. L. A. N., MOSS, A. J., ZAREBA, W., “QT Interval: How to Measure It and what is Normal", Journal of Cardiovascular Electrophysiology, 17(3), 333-336, 2006.
• [14] HELFENBEIN, E. D., ZHOU, S. H., LINDAUER, J. M., FIELD, D. Q., GREGG, R. E., WANG, J. J., KRESQUE, S. S., MICHAUD, F. P., “An Algorithm for Continuous Real-Time QT Interval Monitoring”. Journal of Electrocardiology, 39(4), S123-S127, 2006.
• [15] MURRAY, A., MCLAUGHLIN, N. B., BOURKE, J. P., DOIG, J. C., FURNISS, S. S., CAMPBELL, R. W., “Errors in Manual Measurement of QT Intervals”, British Heart Journal, 71(4), 386-390, 1994.
• [16] MCLAUGHLIN, N. B., CAMPBELL, R. W. F., MURRAY, A., “Accuracy of Automatic QT Measurement Techniques”, Proceedings of Computers in Cardiology, Conference by IEEE, 863-866, London, UK, 1993.
• [17] DASKALOV, I.K., CHRISTOV I.I., "Automatic Detection of the Electrocardiogram T-Wave End", Medical and Biological Engineering and Computing, 37, 348-353, 1999.
• [18] MARTINEZ, J.P., ALMEIDA, R., OLMOS, S., ROCHA, A.P., LAGUNA, P., "A Wavelet-Based ECG Delineator: Evaluation on Standard Databases", IEEE Transcations on Biomedical Engineering, 51(4), 570-581, 2004.
• [19] HAYN, D., KOLLMANN, A., SCHREIER, G., “Automated QT Interval Measurement from Multilead ECG Signals”, IEEE Computers in Cardiology, 381-384, 2006.
• [20] CLIFFORD, G.D., VILLARROEL, M.C., “Model-Based Determination of QT Intervals”, IEEE Computers in Cardiology, 357-360, 2006.
• [21] XUE, Q., REDDY, S., “Algorithms for Computerized QT Analysis”, Journal of Electrocardiology, 30, 181-186, 1998.
• [22] SUAREZ LEON, A.A., MOLINA, D.M., VAZQUEZ SEISDEDOS, C.R., GOOVAERTS, G., VANDEPUT, S., VAN HUFFEL, S., "Neural Network Approach for T Wave End Detection: A comparison of Architectures", Computing in Cardiology, Conference by IEEE, 42, 589-592, Nice, France, 2015.
• [23] İŞCAN, M., YILMAZ, C., YİĞİT, F., "T-Wave End Pattern Classification Based on Gaussian Mixture Model", Signal Processing and Communication Application (SIU), Conference by IEEE, 1953 - 1956, Zonguldak, Türkiye, 2016.
• [24] İŞCAN, M, YİĞİT, F., YILMAZ, C., "Heartbeat Pattern Classification Algorithm Based on Gaussian Mixture Model", International Symposium on Medical Measurements and Applications (MeMeA), Conference by IEEE, 1-6, Benevento, Italy, 2016.
• [25] TSUJI, T., FUKUDA, O., ICHINOBE, H., KANEKO, M., "A log-linearized Gaussian mixture network and its application to EEG pattern classification", IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), 29(1), 60-72, 1999.
• [26] TSUJI, T., BU, N., FUKUDA, O., KANEKO, M., "A Recurrent Log-Linearized Gaussian Mixture Network", IEEE Transactions on Neural Networks, 14(2), 304-316, 2003.
• [27] GOLDBERGER, A.L., LAN, A., GLASS, L., HAUSDORFF, J.M., IVANOV, P.C., MARK, R.G., MIETUS, J.E., MOODY, G.B., PENG, C.K., STANLEY, H.E., “PhysioBank, Physio Toolkit, and PhysioNet: Components of a New Research Resource for Complex Physiological Signals”, Circulation, 101(23), 215-220, 2000.
• [28] YILMAZ, C., İŞCAN, M., YILMAZ, A., “A Fully Automatic Novel Method to Determine QT Interval Based on Continuous Wavelet Transform”, Istanbul University Journal of Electrical and Electronics Engineering, 17(1), 3093-3099, 2017.
• [29] İŞCAN, M., YILMAZ, A., YILMAZ, C., “A Novel Algorithm Combining Continuous Wavelet Transform and Philips Method for QT Interval Analysis”, Electrical, Electronics and Biomedical Engineering (ELECO), National Conference by IEEE, 507-511, Bursa, Türkiye, 2016.