E. Piatkowska-Janko

Computer-aided morphological analysis of Holter ECG recordings based on support vector learning system

The paper presents a new approach to computer-aided analysis of ECG Holter recordings. In contrast to existing tools it is a learning system: the pertinent features of the signal shape are automatically discovered upon the examples carefully selected and commented by cardiologists. Mathematical basis of our system is the theory of support vector machines that are applied for two tasks: signal approximation and pattern classification. Numerical procedures implement the algorithm of sequential minimal optimisation. The computer program is developed in Borland C++ Builder environment. The excellent performances of our approach, high rate of successful pattern recognition and computational efficiency, make use of our tools possible in clinical practice. The system is tested at the Chair and Department of Internal Medicine and Cardiology, Central Teaching Hospital in Warsaw, Poland.

Application of Differential Evolution for Optimization of Least-Square Support Vector Machine Classifier of Signal-Averaged Electrocardiograms

This paper presents the application of differential evolution, an evolutionary algorithm of solving a single objective optimization problem - tuning the hiperparameters of least-square support vector machine classifier. The goal was to improve the classification of patients with sustained ventricular tachycardia after myocardial infarction based on a signal-averaged electrocardiography dataset received from the Medical University of Warsaw. The applied method attained a classification rate of 96% of the SVT+ group.

Detection of atrial electrical instability in hypertensive patients

The aim of our study was to evaluate the influence of the left ventricular hypertrophy (LVH) on the atrial electrical instability in hypertensive patients (pts). 76 hypertensive pts without symptomatic coronary disease, systolic dysfunction, electrolyte disturbances or antiarrhythmic therapy were included in our study, and were divided into two groups: with LVH (group I) 55 pts and without LVH (group II) - 21 pts. Additional measurements were made: ECHO, the 12-lead ECG. From SAECG, the following were calculated: the filtered P-wave duration (hfP); the root mean square voltages for last 30 msec (RMS/sub 30/); RMS voltages for the last time quarter of P-wave divided by RMS voltages of full time of P-wave (RMS4/RMS); envelope of the first 10 ms of P-wave vector magnitude (Penv10). Signal averaged P-wave ECG may be a useful tool to evaluation of atrial electrical instability in hypertensive pts with LVH.

Feature selection of signal-averaged electrocardiograms by orthogonal least squares method

A crucial problem in machine learning is finding the representative set of data for building a model for both classification and approximation task. In this paper we present the orthogonal least squares method for feature selection. The presented method was used for finding the most important features for selecting patients with sustained ventricular tachycardia after myocardial infarction (SVT+). We show that with the reduced set of descriptors used in the classification process we obtain the results that are better than those obtained with the full set.

Semisupervised classifier of signal-average ECG based on k-means clustering

This paper presents the method of risk recognition of sustained ventricular tachycardia and flicker in patients after myocardial infarction based on high-resolution and signal-averaged electrocardiography. Described semisupervised method is combination of k-means clustering and support vector machine classifier. The work is based on dataset obtained from the Medical University of Warsaw. While learning process there were used only 5% examples labels. Evolutionary optimization of coefficients for each signal parameter was executed. It let show the most important parameters. The method of classification had high rate of successful recognition about 94.9%.

Optimal plane search method in blood flow measurements by magnetic resonance imaging

This paper offers an algorithm for determining the blood flow parameters in the neck vessel segments using a single (optimal) measurement plane instead of the usual approach involving four planes orthogonal to the artery axis. This new approach aims at significantly shortening the time required to complete measurements using Nuclear Magnetic Resonance techniques. Based on a defined error function, the algorithm scans the solution space to find the minimum of the error function, and thus to determine a single plane characterized by a minimum measurement error, which allows for an accurate measurement of blood flow in the four carotid arteries. The paper also comprises a practical implementation of this method (as a module of a larger imaging-measuring system), including preliminary research results.