QT interval measurement in cardiac signal processing with multiwavelets Jo¨el Karel∗, Ralf Peeters, Ronald Westra Sandro Haddad, Wouter Serdijn
Electronics Research Laboratory, Mekelweg 4
e-mail: {joel.karel,ralf.peeters,westra}@math.unimaas.nl
{s.haddad,w.a.serdijn}@ewi.tudelft.nl
web page: http://www.math.unimaas.nl/biosens
ABSTRACT
In cardiac research, the measurement of the QT interval duration is important for a number of cardiacpathologies. The QT interval corresponds to the total duration of the ventricular activation, includingboth the depolarization and the repolarization of the ventricles. QT prolongation is considered an indi-cator for sudden cardiac death, see [3]. The QT interval is used to calculate the beat-to-beat variabilityof repolarization (BVR), which is currently an important indicator for some cardiac pathologies, see[5]. The QT interval is also linked to non-cardiac pathologies such as diabetic autonomic neuropathy,see e.g. [1].
The measurement of the QT interval involves the detection of the end-point of the T wave. Asthe T wave is a relatively low-frequency feature in the ECG, low-frequency noise such as baselinewander often hinders accurate detection. In [4] a wavelet-based detection methodology for the QTinterval was introduced. This method first determines the QRS complex and then employs this tosearch for the T peak, using the continuous wavelet transform.
The aim of the current study is to improve on the performance reported in [4] by investigatingthe simultaneous detection of both the location of the QRS complex and the T peak, by usingmultiwavelets from orthogonal filter banks. This builds on the preliminary research reported in [2]. The method employs the discrete wavelet transform which allows for efficient implementation inDSP hardware. The multiwavelets are designed specifically for the task of measuring the QT intervalduration, using the design methodology of [2] and Physionet’s MIT-BIH database. REFERENCES
W.H. Gispen, B. Bravenboer, P.-H. Hendriksen, P.L. Oey, A.C. van Huffelen, and D.W. Erkelens. Is the corrected QT interval a reliable indicator of the severity of diabetic auto- nomic neuropathy? American Diabetes Care, Vol. 16, 1993.
R.L.M. Peeters, J.M.H. Karel, R.L. Westra, S.A.P. Haddad and W.A. Serdijn. MultiwaveletDesign for Cardiac Signal Processing. Proceedings 28th IEEE EMBS Annual InternationalConference, New York City, NY, 2006.
P.E. Puddu and M.G. Bourassa. Prediction of sudden death from QTc interval prolongation in patients with chronic ischemic disease. J. Electrocardiol., Vol. 19, 203–212, 1986.
J.S. Sahambi, S.N. Tandon and R.K.P. Bhatt. An automated approach to beat-by-beat QT- interval analysis. IEEE Eng. Med. Biol. Mag., Vol. 19, 97–101, 2000.
M.B. Thomsen, S.C. Verduyn, M. Stengl, J.D.M. Beekman, G. de Pater, J. van Opstal, P.G.A. Volders and M.A. Vos. Increased Short-Term Variability of Repolarization Predicts d-Sotalol-Induced Torsades de Pointes in Dogs. Circulation, Vol. 110, 2453–2459, 2004.
Student-written practice questions: Midterm 1 1. You are running an experiment in which you want to study the metabolism of E. coli . You know that you have a pure culture because you’ve done a nice series of isolation streaks, and you’re excited to inoculate a new broth culture so you can perform your assays the next day. You perform a series of standard assays including OF medium and mo
This is a description of the results of treatments for a bone fracture of my left foot. The treatments were performed by Hakan Lagergren using the ReeCept X7 laser in Stockholm, Sweden in May 2004. Because I experienced a fracture of the same area in my right foot about ten years ago in New Jersey, U.S.A., it has been interesting to compare the rates of recovery under different treatment. I a