Yalcin Isler

Investigating the performance improvement of HRV Indices in CHF using feature selection methods based on backward elimination and statistical significance

In this study, the best combination of short-term heart rate variability (HRV) measures was investigated to distinguish 29 patients with congestive heart failure from 54 healthy subjects in the control group. In the analysis performed, wavelet packet transform based frequency-domain measures and several non-linear parameters were used in addition to standard HRV measures. The backward elimination and unpaired statistical analysis methods were used to select the best one among all possible combinations of these measures. Five distinct typical classifiers with different parameters were evaluated in discriminating these two groups using the leave-one-out cross validation method. Each algorithm was tested 30 times to determine the repeatability of the results. The results imply that the backward elimination method gives better performance when compared to the statistical significance method in the feature selection stage. The best performance (82.75%, 96.29%, and 91.56% for the sensitivity, specificity, and accuracy) was obtained by using the SVM classifier with 27 selected features including non-linear and wavelet-based measures.

Heart rate normalization in the analysis of heart rate variability in congestive heart failure.

Abstract In this study, the effects of heart rate (HR) normalization in the analysis of the heart rate variability (HRV) were investigated to distinguish 29 patients with congestive heart failure from 54 healthy subjects in the control group. In the analysis performed, the best accuracy performances of optimal combination of standard short-term HRV measures and of HR-normalized short-term HRV measures are compared. A genetic algorithm is used to select the best features from among all possible combinations of these measures. A k-nearest-neighbour (KNN) classifier is used to evaluate the performances of the feature combinations in classifying these two data groups. The results imply that using both min—max and HR normalization improves the performance of the classification. The maximum accuracy is achieved as 93.98 per cent using k = 3 and k = 5 for the KNN classifier with the perfect positive predictivity values.

Comparison of the Effects of Cross-validation Methods on Determining Performances of Classifiers Used in Diagnosing Congestive Heart Failure

Abstract Congestive heart failure (CHF) occurs when the heart is unable to provide sufficient pump action to maintain blood flow to meet the needs of the body. Early diagnosis is important since the mortality rate of the patients with CHF is very high. There are different validation methods to measure performances of classifier algorithms designed for this purpose. In this study, k-fold and leave-one-out cross-validation methods were tested for performance measures of five distinct classifiers in the diagnosis of the patients with CHF. Each algorithm was run 100 times and the average and the standard deviation of classifier performances were recorded. As a result, it was observed that average performance was enhanced and the variability of performances was decreased when the number of data sections used in the cross-validation method was increased.

A computer software for simulating single-compartmental model of neurons

In this paper, a new computer software package, Yalzer, is introduced for simulating single-compartmental model of neurons. Passive or excitable membranes with voltage-gated ion channels can be modeled, and current clamp and voltage clamp experiments can be simulated. In the Yalzer, first-order differential equations used to define the dynamics of the gate variables and the membrane potential are solved by two separate integration methods with variable time steps: forward Euler and exponential Euler methods. Outputs of the simulation are shown on a spreadsheet template for allowing flexible data manipulation and can be graphically displayed. The user can define the model in detail, and examine the excitability of the model and the dynamics of voltage-gated ion channels. The software package addresses to ones who want to run simple simulations of neurons without need to any programming language skills or expensive software. It can also be used for educational purposes.

Discrimination of systolic and diastolic dysfunctions using multi-layer perceptron in heart rate variability analysis

In this study, the heart rate variability (HRV) analysis is used to distinguish patients with systolic congestive heart failure (CHF) from patients with diastolic CHF. In the analysis performed, the best accuracy performances of short-term HRV measures are compared. These measures are calculated in four different ways with optional normalization methods of heart rate and data. The nearest neighbor and the multi-layer perceptron (MLP) are used to evaluate the performances in discriminating these two groups. The results point out that using both data and heart rate normalizations enhances the classifier performance. The maximum accuracy is obtained as 96.43% with MLP classifier.

Operation of Compressor and Electronic Expansion Valve via Different Controllers

The most critical problem in the world is to meet the energy demand, because of steadily increasing energy consumption. Refrigeration systems` electricity consumption has big portion in overall consumption. Therefore, considerable attention has been given to refrigeration capacity modulation system in order to decrease electricity consumption of these systems. Capacity modulation is used to meet exact amount of load at partial load and lowered electricity consumption by avoiding over capacity using. Variable speed refrigeration systems are the most common capacity modulation method for commercially and household purposes. Although the vapor compression refrigeration designed to satisfy the maximum load, they work at partial load conditions most of their life cycle and they are generally regulated as on/off controlled. The experimental chiller system contains four main components: compressor, condenser, expansion device, and evaporator in Fig.1 where this study deals with effects of different control methods on variable speed compressor (VSC) and electronic expansion valve (EEV). This chiller system has a scroll type VSC and a stepper motor controlled EEV.

A microcontroller based test platform for controller design

This paper presents a microcontroller based emulator which is developed for the design-test-and-redesign of controllers through examining the controller candidates on the emulated plants operating under the created real-time and real environmental conditions. The analog and digital peripheral units of the plant emulator provide the recreation of the real environmental conditions such that the whole emulator runs as if the real plant whenever the model implemented in the microcontroller is a right model for the plant. The emulator has the flexibility of changing the plant model and/or its parameters manually and it also enables to apply, in an automatic way, analog disturbances and analog parameter perturbations to the plant model. The developed plant emulator communicated with the simulator of the controller part both managed by a graphical user interface implemented in the computer constitutes a controller design-test-redesign platform which has the advantages of using simulated plants in the controller design providing a test-flexibility and also of testing the controller candidates on the emulated plants operating under the conditions very close to the real situations, so assuring the desired controller performances.

Investigating the Effects of Wavelet Entropy in Heart Rate Variability Analysis for Diagnosing of Congestive Heart Failure

In this study, wavelet entropy, which is calculated from the wavelet transform coefficients obtained from heart rate variability data, is used to distinguish the control group from the patients with congestive heart failure. Wavelet entropies are obtained from 29 patients with congestive heart failure and 54 subjects in the control group. Standard heart rate variability measurements are also calculated for the whole dataset. Finally, linear discriminant analysis is used to evaluate the performance of measurements in classifying these two groups

Effect of linear and non-linear measurements of heart rate variability in prediction of PAF attack

Paroxysmal Atrial Fibrillation (PAF) is a very common rhythm disorder that causes rapid and irregular impulses in the heart. In this study, it is aimed to determine whether patients can be warned before PAF events. 30-minute HRV data used in this study. Each piece of data was divided into 10 pieces of 5-minute parts. Time domain measurements from linear measurements of HRV and Poincare measurements from nonlinear measurements of HRV were used for each segment. Detecting performances were measured for each segment using k-nearest neighbor classifier. Particularly linear measurements have been shown to achieve up to 82% success in predicting PAF attack and was observed that PAF attack could be detected 12,5 minutes earlier.

Early prediction of Paroxysmal Atrial Fibrillation using frequency domain measures of heart rate variability

Paroxysmal Atrial Fibrillation (PAF) is a very common heart disease caused by irregular impulses of atrial tissue in adult. Diagnosing in the early stages of this disorder is very important for the patients to stop the progression of the disease and to improve the life quality. In this study, it is aimed to predict the PAF event before the realization of the PAF which in 5 minutes for the PAF patients. 30-minute data used in the study were divided into 5-minute parts. Fast Fourier Transform of frequency domain measures of heart rate variability obtained easily and practically is used for each part. The statistical significances among segments and discriminating performances of k-Nearest Neighbors classifier were obtained for each segment using these measurements. Consequently, As a result of statistical analysis, it is shown that patients may be warned 12.5 minutes earlier than a PAF attack.