Konstantin E. Nemchenko
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Konstantin E. Nemchenko has a total of 95 co-authors. Here are the top co-authors by number of shared publications:
I. N. Adamenko
V. N. Karazin Kharkiv National University , Kharkiv 61022, Ukraine
32 shared publicationsA. F. G. Wyatt
University of Exeter 3 School of Physics Exeter EX4 4QL, United Kingdom
16 shared publications- A. F. G. Wyatt
School of Physics, University of Exeter, Exeter EX4 4QL, United Kingdom
11 shared publications - T. Vikhtinskaya
V.N. Karazin Kharkiv National University, Kharkiv, Ukraine
10 shared publications - I. N. Adamenko
Kharkov State University, 4 Svobody sq., 310077 Kharkov, Ukraine
9 shared publications V. A. Slipko
V.N. Karazin Kharkov National University
9 shared publications
Latest Publications
- Open Access
- 25 June 2026
- Published by V. N. Karazin Kharkiv National University in Biophysical Bulletin
- No. 55,p. 117-129
Improving the efficiency of spine region segmentation using an ensemble of pre-trained neural networks
Background: The accuracy of segmentation of vertebrae in X-ray images is critical for clinical decisions as the manual method is laborious. The use of deep learning is complicated by low contrast, noise, and patient position artifacts. These negative factors make a single neural network unreliable. Thus, to improve the accuracy and efficiency of segmentation, regardless of the quality of X-ray images, there is a need for an ensemble of neural networks that compensates for the individual shortcomings of the models by aggregating their results. Objectives: Increasing the accuracy and efficiency of segmentation of a spinal region consisting of four vertebrae (Th8, Th9, Th10, Th11) in X-ray images by using an ensemble of pre-trained neural networks. Materials and methods: Two datasets were used for the experiments: the first set with 183 images was distributed in the ratio of 70% / 10% / 20% for training, validation, and testing, in turn, the second set of 58 images was used exclusively for the final assessment of the generalization ability of the ensemble on new data. In the process of research, segmentation accuracy with and without augmentation was first compared, after which the 10 best from the initial 20 neural networks were selected for further use, and five ensemble algorithms were used for mask aggregation. Results: For the ensemble of pre-trained neural networks, the best result was shown by soft voting. Comparing the obtained result with the results presented by Koniukhov et al. (2024), the improvement was 3.06%. This indicator clearly confirms the effectiveness of using pre-trained networks for segmentation of the spine area. Conclusions: Soft voting for an ensemble of pre-trained neural networks demonstrated the greatest improvement in segmentation accuracy compared to other methods. Aggregating knowledge from 10 models successfully eliminated the limitations of individual models. The use of an ensemble of pre-trained neural networks improved segmentation accuracy for both the test data from the first dataset and the data from the second dataset. Such results confirm the feasibility of applying the proposed ensemble-based approach to chest X-ray radiographs for vertebrae segmentation in medical imaging tasks.
- 25 June 2026
- Published by V. N. Karazin Kharkiv National University in Biophysical Bulletin
- No. 55,p. 117-129
Open AccessImproving the efficiency of spine region segmentation using an ensemble of pre-trained neural networks
Background: The accuracy of segmentation of vertebrae in X-ray images is critical for clinical decisions as the manual method is laborious. The use of deep learning is complicated by low contrast, noise, and patient position artifacts. These negative factors make a single neural network unreliable. Thus, to improve the accuracy and efficiency of segmentation, regardless of the quality of X-ray images, there is a need for an ensemble of neural networks that compensates for the individual shortcomings of the models by aggregating their results. Objectives: Increasing the accuracy and efficiency of segmentation of a spinal region consisting of four vertebrae (Th8, Th9, Th10, Th11) in X-ray images by using an ensemble of pre-trained neural networks. Materials and methods: Two datasets were used for the experiments: the first set with 183 images was distributed in the ratio of 70% / 10% / 20% for training, validation, and testing, in turn, the second set of 58 images was used exclusively for the final assessment of the generalization ability of the ensemble on new data. In the process of research, segmentation accuracy with and without augmentation was first compared, after which the 10 best from the initial 20 neural networks were selected for further use, and five ensemble algorithms were used for mask aggregation. Results: For the ensemble of pre-trained neural networks, the best result was shown by soft voting. Comparing the obtained result with the results presented by Koniukhov et al. (2024), the improvement was 3.06%. This indicator clearly confirms the effectiveness of using pre-trained networks for segmentation of the spine area. Conclusions: Soft voting for an ensemble of pre-trained neural networks demonstrated the greatest improvement in segmentation accuracy compared to other methods. Aggregating knowledge from 10 models successfully eliminated the limitations of individual models. The use of an ensemble of pre-trained neural networks improved segmentation accuracy for both the test data from the first dataset and the data from the second dataset. Such results confirm the feasibility of applying the proposed ensemble-based approach to chest X-ray radiographs for vertebrae segmentation in medical imaging tasks.
- Open Access
- 1 January 2025
- Published by AIP Publishing in Low Temperature Physics
- Vol. 51 (1) , 5-9
Two-mode dissipation of oscillating tuning fork in 3He–4He superfluid mixtures
In this paper, we present the results of theoretical studies of the possible mechanism of energy dissipation that takes place in the experiments with vibrating quartz fork in superfluid 3He–4He mixtures with high concentration. The possibility and efficiency of the tuning fork radiation of both the second sound wave and the diffusion dissipative wave were investigated. The relationships between the amplitudes of temperature and concentration oscillations in these waves were found and the relative wave intensities were determined. This allowed us to give a possible explanation for observed additional mechanisms of tuning fork attenuation in helium solutions.
- 1 January 2025
- Published by AIP Publishing in Low Temperature Physics
- Vol. 51 (1) , 5-9
Open AccessTwo-mode dissipation of oscillating tuning fork in 3He–4He superfluid mixtures
In this paper, we present the results of theoretical studies of the possible mechanism of energy dissipation that takes place in the experiments with vibrating quartz fork in superfluid 3He–4He mixtures with high concentration. The possibility and efficiency of the tuning fork radiation of both the second sound wave and the diffusion dissipative wave were investigated. The relationships between the amplitudes of temperature and concentration oscillations in these waves were found and the relative wave intensities were determined. This allowed us to give a possible explanation for observed additional mechanisms of tuning fork attenuation in helium solutions.
- Open Access
- 8 August 2024
- Published by V. N. Karazin Kharkiv National University in Biophysical Bulletin
- No. 51,p. 39-52
Segmented algorithm for three-dimensional reconstruction in linear scan geometry
Background: Computer tomography is recognized as one of the most powerful methods for diagnosis, and monitoring of a wide range of diseases. It provides the ability to obtain detailed information about the internal structure of organs and bones. Despite the success of computed tomography in areas such as three-dimensional mammography or lung radiography, it has not achieved the same level of widespread as, for example, magnetic resonance imaging, even if CT offers greater accuracy. This is primarily due to safety limitations on the permissible number of examinations due to the harmfulness of X-ray radiation to the patient. One of the main challenges facing researchers is the need to reduce the time of the entire examination and decrease the radiation exposure to the patient. Overcoming these challenges is crucial for improving the overall efficiency of medical services, optimizing treatment plans, and ultimately enhancing patient outcomes. Thus, addressing these issues through innovative algorithms and methods in computed tomography holds significant potential for revolutionizing medical diagnostics and ensuring continuous progress in modern healthcare. Objectives: The aim of this work was to develop an algorithm for three-dimensional reconstruction that is independent of the conicity of the radiation beam. Therefore, it allows for accurate reconstruction of the entire object with a single rotation of the radiation source around the investigated object. Materials and Methods: The work utilizes methods of integral transforms and computer modeling to solve inverse problems arising in computer tomography. Results: An analytical inversion formula was obtained for three-dimensional computer tomography with linear scan geometry and segmentation. The feasibility of the developed algorithm was verified, and a methodology for research with linear motion of the conical emitter and detectors was developed. Conclusions: The developed algorithm improves the reconstruction of object layers significantly distant from the plane in which the emitter and detector move, compared to existing algorithms.
- 8 August 2024
- Published by V. N. Karazin Kharkiv National University in Biophysical Bulletin
- No. 51,p. 39-52
Open AccessSegmented algorithm for three-dimensional reconstruction in linear scan geometry
Background: Computer tomography is recognized as one of the most powerful methods for diagnosis, and monitoring of a wide range of diseases. It provides the ability to obtain detailed information about the internal structure of organs and bones. Despite the success of computed tomography in areas such as three-dimensional mammography or lung radiography, it has not achieved the same level of widespread as, for example, magnetic resonance imaging, even if CT offers greater accuracy. This is primarily due to safety limitations on the permissible number of examinations due to the harmfulness of X-ray radiation to the patient. One of the main challenges facing researchers is the need to reduce the time of the entire examination and decrease the radiation exposure to the patient. Overcoming these challenges is crucial for improving the overall efficiency of medical services, optimizing treatment plans, and ultimately enhancing patient outcomes. Thus, addressing these issues through innovative algorithms and methods in computed tomography holds significant potential for revolutionizing medical diagnostics and ensuring continuous progress in modern healthcare. Objectives: The aim of this work was to develop an algorithm for three-dimensional reconstruction that is independent of the conicity of the radiation beam. Therefore, it allows for accurate reconstruction of the entire object with a single rotation of the radiation source around the investigated object. Materials and Methods: The work utilizes methods of integral transforms and computer modeling to solve inverse problems arising in computer tomography. Results: An analytical inversion formula was obtained for three-dimensional computer tomography with linear scan geometry and segmentation. The feasibility of the developed algorithm was verified, and a methodology for research with linear motion of the conical emitter and detectors was developed. Conclusions: The developed algorithm improves the reconstruction of object layers significantly distant from the plane in which the emitter and detector move, compared to existing algorithms.
- Open Access
- 21 June 2024
- Published by V. N. Karazin Kharkiv National University in Bulletin of V.N. Karazin Kharkiv National University, series «Mathematical modeling. Information technology. Automated control systems»
- Vol. 62 (62) , 37-44
Multiple training of neural networks for automatic spine segmentation
Preventive and diagnostic studies of the presence of bone diseases require morphometric studies of X-ray images of the chest area. Nowadays, artificial intelligence methods are increasingly being used to solve such problems. The main difficulties of this task are related to the fact that X-ray images have quality limitations, for example, in terms of signal-to-noise ratio or contrast. For this reason, the application of standard methods of image recognition or automatic diagnosis becomes impossible. These difficulties have led to the fact that there is currently a fairly large number of works in this field, but the results of most of them are insufficient for practical use. Goal Investigate the possibility of using artificial intelligence in the segmentation of medical images for the purpose of automatic diagnosis of diseases of the human bone system. Research methods The ensemble method of X-ray image segmentation has been used in the study. The baseline of training data was created on the basis of X-ray images taken from open sources. The total number of images is 183. The initial data was modified according to the requirements necessary for model training. All images were converted to grayscale and resized to 256x256 pixels. Results Using this method in the two test cases resulted in an improvement in accuracy from 0.543 to 0.820 for the first snapshot and from 0.725 to 0.923 for the second snapshot. Conclusions We have proposed and investigated the application of the methodology of using an ensemble of reusable neural networks for automatic segmentation of a certain area of the spine, namely the Th8-Th11 spine region. The application of this method allowed obtaining more stable and accurate predictions for the desired spine regions, even for images with high noise levels.
- 21 June 2024
- Published by V. N. Karazin Kharkiv National University in Bulletin of V.N. Karazin Kharkiv National University, series «Mathematical modeling. Information technology. Automated control systems»
- Vol. 62 (62) , 37-44
Open AccessMultiple training of neural networks for automatic spine segmentation
Preventive and diagnostic studies of the presence of bone diseases require morphometric studies of X-ray images of the chest area. Nowadays, artificial intelligence methods are increasingly being used to solve such problems. The main difficulties of this task are related to the fact that X-ray images have quality limitations, for example, in terms of signal-to-noise ratio or contrast. For this reason, the application of standard methods of image recognition or automatic diagnosis becomes impossible. These difficulties have led to the fact that there is currently a fairly large number of works in this field, but the results of most of them are insufficient for practical use. Goal Investigate the possibility of using artificial intelligence in the segmentation of medical images for the purpose of automatic diagnosis of diseases of the human bone system. Research methods The ensemble method of X-ray image segmentation has been used in the study. The baseline of training data was created on the basis of X-ray images taken from open sources. The total number of images is 183. The initial data was modified according to the requirements necessary for model training. All images were converted to grayscale and resized to 256x256 pixels. Results Using this method in the two test cases resulted in an improvement in accuracy from 0.543 to 0.820 for the first snapshot and from 0.725 to 0.923 for the second snapshot. Conclusions We have proposed and investigated the application of the methodology of using an ensemble of reusable neural networks for automatic segmentation of a certain area of the spine, namely the Th8-Th11 spine region. The application of this method allowed obtaining more stable and accurate predictions for the desired spine regions, even for images with high noise levels.
- 1 May 2024
- Published by AIP Publishing in Low Temperature Physics
- Vol. 50 (5) , 364-367
Unusual heat transfer dependence on boundary scattering in very long narrow conductors
When modeling heat transfer in nanoribbons, an unusually sharp dependence of the heat flux, and, consequently, the thermal conductivity coefficient, on the degree of specularity of phonon reflection from the lateral boundaries was found. From numerical calculations, it followed that the addition of minimally diffuse scattering should lead to an abrupt increase in the thermal resistance of the conductor. Given the importance of selecting optimal thermal parameters in nanotechnology, this short paper provides an analysis of the found dependencies, discusses the physical reasons for this anomalous behavior, and draws conclusions about the importance of achieving the quality of the nanoconductor interface to create thermal contacts with a given or optimal thermal conductivity.
- 1 May 2024
- Published by AIP Publishing in Low Temperature Physics
- Vol. 50 (5) , 364-367
Unusual heat transfer dependence on boundary scattering in very long narrow conductors
When modeling heat transfer in nanoribbons, an unusually sharp dependence of the heat flux, and, consequently, the thermal conductivity coefficient, on the degree of specularity of phonon reflection from the lateral boundaries was found. From numerical calculations, it followed that the addition of minimally diffuse scattering should lead to an abrupt increase in the thermal resistance of the conductor. Given the importance of selecting optimal thermal parameters in nanotechnology, this short paper provides an analysis of the found dependencies, discusses the physical reasons for this anomalous behavior, and draws conclusions about the importance of achieving the quality of the nanoconductor interface to create thermal contacts with a given or optimal thermal conductivity.
- Open Access
- 1 January 2024
- Published by Taras Shevchenko National University of Kyiv in Medical physics – the current status, problems, the way of development. Innovation technologies
- No. 1,p. 329-334
MODIFIED LINEAR SCANNING ALGORITHMS IN COMPUTED TOMOGRAPHY
Introduction. This paper investigates and analyzes the effectiveness of a new approach to computed tomography, based on cone-beam tomography using a linear trajectory of the radiation source. Standard approximate reconstruction methods, despite their speed, can lead to a decrease in image quality, which is critical for diagnostics, such as detecting small-sized tumors at early stages. Methods. The proposed approach is based on exact reconstruction methods, where the projection data obtained from such scanning geometry is complete, allowing for an accurate and high-quality reconstruction of the object. However, the practical application of such scanning geometry has certain drawbacks and limitations, so several possible modifications of the algorithm are proposed. Results. Additionally, the paper presents the results of a study on the effectiveness of image reconstruction methods based on simulations of different scanning scenarios. The results indicate high reconstruction accuracy when using a square-closed trajectory with a limited path length of the radiation source. Conclusions. An analytical reconstruction algorithm was developed and compared with the classical Feldkamp-Davis-Kress (FDK) algorithm. The results showed that the new algorithm provides more accurate reconstruction and faster data processing, which will contribute to improving diagnostic procedures in medical practice.
- 1 January 2024
- Published by Taras Shevchenko National University of Kyiv in Medical physics – the current status, problems, the way of development. Innovation technologies
- No. 1,p. 329-334
Open AccessMODIFIED LINEAR SCANNING ALGORITHMS IN COMPUTED TOMOGRAPHY
Introduction. This paper investigates and analyzes the effectiveness of a new approach to computed tomography, based on cone-beam tomography using a linear trajectory of the radiation source. Standard approximate reconstruction methods, despite their speed, can lead to a decrease in image quality, which is critical for diagnostics, such as detecting small-sized tumors at early stages. Methods. The proposed approach is based on exact reconstruction methods, where the projection data obtained from such scanning geometry is complete, allowing for an accurate and high-quality reconstruction of the object. However, the practical application of such scanning geometry has certain drawbacks and limitations, so several possible modifications of the algorithm are proposed. Results. Additionally, the paper presents the results of a study on the effectiveness of image reconstruction methods based on simulations of different scanning scenarios. The results indicate high reconstruction accuracy when using a square-closed trajectory with a limited path length of the radiation source. Conclusions. An analytical reconstruction algorithm was developed and compared with the classical Feldkamp-Davis-Kress (FDK) algorithm. The results showed that the new algorithm provides more accurate reconstruction and faster data processing, which will contribute to improving diagnostic procedures in medical practice.
- Open Access
- 1 August 2023
- Published by AIP Publishing in Low Temperature Physics
- Vol. 49 (8) , 961-966
Analytical description of phonon heat flow with diffusive boundary scattering
The paper presents an analytical solution to the problem of heat flow due to the ballistic motion of phonons, which are diffusely reflected from the side walls of a two-dimensional dielectric material. An explicit analytical expression is obtained for the temperature profile in the approximation of linear dependence; explicit equations are derived for the heat flux and the thermal conductivity coefficients of the sample with arbitrary ratios of the width W and the length L. The analytical expressions are in very good agreement with numerical solutions of the integral equations, which describe these quantities and were derived in our previous study [J. Appl. Phys. 129, 085105 (2021)]. The obtained analytical relations can be used for subsequent studies in the case of mixed specular and diffuse reflection, as well as in the study of phonon systems of specific materials.
- 1 August 2023
- Published by AIP Publishing in Low Temperature Physics
- Vol. 49 (8) , 961-966
Open AccessAnalytical description of phonon heat flow with diffusive boundary scattering
The paper presents an analytical solution to the problem of heat flow due to the ballistic motion of phonons, which are diffusely reflected from the side walls of a two-dimensional dielectric material. An explicit analytical expression is obtained for the temperature profile in the approximation of linear dependence; explicit equations are derived for the heat flux and the thermal conductivity coefficients of the sample with arbitrary ratios of the width W and the length L. The analytical expressions are in very good agreement with numerical solutions of the integral equations, which describe these quantities and were derived in our previous study [J. Appl. Phys. 129, 085105 (2021)]. The obtained analytical relations can be used for subsequent studies in the case of mixed specular and diffuse reflection, as well as in the study of phonon systems of specific materials.
- 1 April 2023
- Published by AIP Publishing in Low Temperature Physics
- Vol. 49 (4) , 428-432
Creation and evolution of a “hotline” in superfluid helium
The paper provides a theoretical consideration of the “hotline” phenomenon, a region with an increased phonon energy density, which occurs when two beams cross in helium. An analysis of the results of observations and a theoretical consideration of the formation of a “hotline” are carried out. The results of calculations are also compared with experimental data on the dependence of the “hotline” parameters on the power of the initial beams, and the angle between the directions of their movement.
- 1 April 2023
- Published by AIP Publishing in Low Temperature Physics
- Vol. 49 (4) , 428-432
Creation and evolution of a “hotline” in superfluid helium
The paper provides a theoretical consideration of the “hotline” phenomenon, a region with an increased phonon energy density, which occurs when two beams cross in helium. An analysis of the results of observations and a theoretical consideration of the formation of a “hotline” are carried out. The results of calculations are also compared with experimental data on the dependence of the “hotline” parameters on the power of the initial beams, and the angle between the directions of their movement.
- 1 February 2023
- Published by AIP Publishing in Low Temperature Physics
- Vol. 49 (2) , 171-174
Collective mode resonances in superfluid 3He–4He mixtures
The experimentally observed first and second sound resonances caused by vibrations of a closed quartz tuning fork in superfluid helium mixtures are studied. Taking into account the full set of hydrodynamic modes of the system allows us to explain the unusual behavior of second sound resonances observed in experiments due to the redistribution of energy between the dissipative mode and the second sound mode.
- 1 February 2023
- Published by AIP Publishing in Low Temperature Physics
- Vol. 49 (2) , 171-174
Collective mode resonances in superfluid 3He–4He mixtures
The experimentally observed first and second sound resonances caused by vibrations of a closed quartz tuning fork in superfluid helium mixtures are studied. Taking into account the full set of hydrodynamic modes of the system allows us to explain the unusual behavior of second sound resonances observed in experiments due to the redistribution of energy between the dissipative mode and the second sound mode.
- 1 August 2022
- Published by AIP Publishing in Low Temperature Physics
- Vol. 48 (8) , 583-588
Kinetic properties of solid 3He–4He mixtures in the model of freely moving impuritons
Within the framework of the gas-kinetic model of a mixture of weakly interacting gas of quasiparticles—phonons and impurities—the dissipative properties of 3He–4He solid solutions are considered: diffusion, self-diffusion, and thermal conductivity in the range of concentrations and temperatures, where impurities of 3He can be considered as quasiparticles freely moving through the crystal. Explicit expressions are obtained for the dissipative coefficients of spin and mass diffusion and the coefficient of thermal conductivity. The observed in experiments on spin relaxation, heat transfer, and separation of solutions dependencies and parameters of quasiparticles are discussed.
- 1 August 2022
- Published by AIP Publishing in Low Temperature Physics
- Vol. 48 (8) , 583-588
Kinetic properties of solid 3He–4He mixtures in the model of freely moving impuritons
Within the framework of the gas-kinetic model of a mixture of weakly interacting gas of quasiparticles—phonons and impurities—the dissipative properties of 3He–4He solid solutions are considered: diffusion, self-diffusion, and thermal conductivity in the range of concentrations and temperatures, where impurities of 3He can be considered as quasiparticles freely moving through the crystal. Explicit expressions are obtained for the dissipative coefficients of spin and mass diffusion and the coefficient of thermal conductivity. The observed in experiments on spin relaxation, heat transfer, and separation of solutions dependencies and parameters of quasiparticles are discussed.