Перегляд за автором "Pulavskyi, A. A."
Зараз показано 1 - 8 з 8
Результатів на сторінку
Варіанти сортування
Публікація Assessment of Electrocardiogram Quality Using Lossless Compression Technique for Heart Rate Variability Analysis(IEEE, 2019) Кривенко, С. А.; Pulavskyi, A. A.; Kolesnikova, O. V.; Lukin, V. V.; Posokhov, M. F.; Krivenko, S. S.A simple method of assessment of electrocardiogram (ECG) quality based on lossless data compression is proposed. It is intended for a quick preliminary assessment of ECG in terms of further analysis of heart rate variability (HRV) based on a sequence of R-peaks. The method does not require a priori information about the type, nature and intensity of noise. The only output parameter of pre-processing step is the ECG compression ratio. Thresholds of compression ratio are determined, above which the results of further analysis of HRV are the same as for the corresponding ECG not distorted by noise. For a sampling frequency of 250 Hz, the threshold is equal to 1.4, for a frequency of 2000 Hz, it is 4.0. The results are shown for both synthetic ECG and ECG received using a personal screening diagnostic device.Публікація Automatic recognition of congestive heart failure signs in heart rate variability data(IEEE, 2022) Pulavskyi, A. A.; Krivenko, S. S.; Krivenko, S. A.; Linskiy, I. V.; Posokhov, M. F.; Kryvenko, L. S.Automatic screening of the population for congestive heart failure (CHF) is a matter of pressing concern due to the severity of the health consequences resulting in disability and death of people. On the one hand, portable devices working with ECG signals become convenient tools for the lay user due to the simplicity. On the other hand, analyzing the specific behavior of the R-peaks sequence (analysis of heart rate variability) in cardiac pathologies allows identifying the patterns inherent in particular heart dysfunction. Such patterns are effectively differentiated using symbolic dynamics methods and the subsequent application of machine learning methods. In this study, a highly specific model was obtained (sensitivity 0.71, specificity 0.96), suitable for automatic screening of CHF. Its operability and performance characteristics have been verified through testing in several publicly available databases.Публікація Determination of low hemoglobin level in human using the analysis of symbolic dynamics of the heart rate variability,(IEEE, 2017) Кривенко, С. А.; Krivenko, S. S.; Pulavskyi, A. A.Symbolic dynamics of electrocardiograms (ECG) carries the information about functioning of various human body systems. A method for distinguishing the men with low hemoglobin value from the men with normal hemoglobin value by analyzing the symbolic dynamics of the heart rate variability was proposed in the research. The method has got an acceptable sensitivity (0.67-0.80) and specificity (0.80-1.00) and works by using a single lead ECG.Публікація Identification of Diabetic Patients Using the Nonlinear Analysis of Short-Term Heart Rate Time Series(IEEE, 2018) Кривенко, С. А.; Krivenko, S. S.; Pulavskyi, A. A.The non-invasive method for identifying the volunteers suffering from the type 2 diabetes mellitus (T2DM) is suggested. The method is based on the symbolic analysis of short series (~300 points) of RR-intervals of a single-lead electrocardiogram. To obtain the initial symbol sequences, 4 different methods of formation of symbols from the time series were used. Using the SVM classifier with a linear kernel, a selection of significant symbols was made. The most significant symbols became the input parameters for the SVM model with RBF kernel. The model has shown high efficiency: the sensitivity on the test sets was 70-82%, and the specificity was 73-77%. The proposed method uses the posterior probability, which has been accompanied by a class label for each new sample, being a criterion of the results reliability. For the obtained model, the threshold value of the posterior probability was 91%. It has been shown that the use of the posterior probability does not impair or improve the quality of the forecast. While using the posterior probability, the sensitivity of the model can increase up to 88% and specificity can increase up to 90%, being objective for up to 50% of all predicted values.Публікація Many-To-Many Linear-Feedback Shift Model for Training of Artificial Neural Network in Dentistry(IEEE, 2019) Кривенко, С. А.; Pulavskyi, A. A.; Krivenko, S. S.; Kryvenko, L. S.In this paper, the authors consider how to label and save a large number of images that should be predict in a single file. Technique of automatic labeling the data set with finite element model for training of artificial neural network in tomography are proposed. Simple transparent example of sixteen images for predict in a single HDF5 file training of artificial neural network in tomography show accuracy 100% for training set as well for test set. Then this technique is able to build information model of salivary immune and periodontal status and to evaluate the correlation between salivary immunoglobulin level, inflammation in periodontal tissues and orthodontic pathology. The study was conducted on 139 subjects, which were in the age group of 12-18 years reporting to the Department of Pediatric Dentistry of Kharkiv National Medical University. The atopic group consisted of 103 patients with the following conditions: 76 patients of atopic diseases and gingivitis (Group 1) and 27 patients of atopic diseases, gingivitis and orthodontic pathology (Group 2). Among the 139 subjects, 36 healthy controls formed Group 3. The obtained data prove that there is an immune misbalance in children with atopy and in children with combined atopic and orthodontic pathology. Level of sIgA and IgG is decreased in group of patients with atopy and in group of children with atopic and orthodontic pathology. The information model of salivary immune and periodontal status was built and regression analysis showed that there was strong correlation between inflammation in periodontal tissues and level of immunoglobulins.Публікація Noninvasive evaluation of glucose concentration in the human blood based on electrocardiograms(IEEE, 2015) Кривенко, С. А.; Pulavskyi, A. A.; Krivenko, S. S.A method for detection of twofold or greater change in the concentration of glucose in the blood of patients suffering from type 2 diabetes by means of the analysis of their electrocardiograms is proposed. The method is based on a tenth order linear prediction filter. The square of the Matthew's correlation coefficient 0.91 has been achieved during the examination of 21 patients by means of the standard oral glucose tolerance test with the proposed method.Публікація The computation of line spectral frequencies using discrete wavelet transform for electrocardiograms processing(IEEE, 2016) Кривенко, С. А.; Pulavskyi, A. A.; Krivenko, S. S.An approach for glucose meters was improved. We process electrocardiograms of patients to detect change of glucose concentration in the blood. The technique uses discrete wavelet transform for electrocardiograms processing. Standard oral glucose tolerance test used to check fifty six patients complains of type 2 diabetes symptoms. The square of the Matthews correlation coefficient 0.93 has been achieved on examination.Публікація The use of lossless compression in the process of post-filtration smoothing of an ECG distorted by high muscle noise(IEEE, 2020) Кривенко, С. А.; Pulavskyi, A. A.; Kryvenko, L. S.; Posokhov, M. F.; Krivenko, S. S.The electrocardiogram (ECG) of the first lead, obtained using portable signal collection tools is noise sensitive. One of the most common and poorly filterable types of noise is muscle artefact. As a result of the simulation, a criterion was found that identifies situations in which the use of kernel-based ECG smoothing after the filtering stage is relevant. This criterion is based on the difference in compression coefficients of the ECG smoothed after filtering and filtered ECG. The use of post-filtering smoothing is recommended if this difference is greater than zero. In this case, post-filtering smoothing improves signal quality in terms of mean-square error. The correctness of the criterion was confirmed on real ECG results (MIT-BIH Arrhythmia Database), distorted by real muscle artefact (MIT-BIH Noise Stress Test Database).