Institution
Odessa National Polytechnic University
Education•Odesa, Ukraine•
About: Odessa National Polytechnic University is a education organization based out in Odesa, Ukraine. It is known for research contribution in the topics: Computer science & Grinding. The organization has 806 authors who have published 677 publications receiving 2504 citations.
Papers published on a yearly basis
Papers
More filters
••
TL;DR: It is demonstrated that contrary to the common perception of QSAR models as "black boxes" they can be used to identify statistically significant chemical substructures (QSAR-based alerts) that influence toxicity.
90 citations
••
Abstract: In this paper, we consider dynamical behavior of astrophysical objects such as galaxies and dwarf galaxies taking into account both the gravitational attraction between them and the cosmological expansion of the Universe. First, we obtain the general system of equations and apply them to some abstract systems of galaxies. Then we investigate the collision between the Milky Way and Andromeda in future. Here, we distinguish two models. For the first one, we do not take into account the influence of the Intra-Group Matter (IGrM). In this case, we demonstrate that for currently known parameters of this system the collision is hardly plausible because of the angular momentum. These galaxies will approach the minimum distance of about 290 Kpc in 4.44 Gyr from present, and then begin to run away irreversibly from each other. For the second model, we take into account the dynamical friction due to the IGrM. Here, we find a characteristic value of the IGrM particle velocity dispersion σ-tilde = 2.306. For σ-tilde ≤ 2.306, the merger will take place, but for the bigger values of σ-tilde the merger can be problematic. If the temperature of the IGrM particles is 10{sup 5} K, then this characteristic value ofmore » σ-tilde corresponds to the IGrM particle mass 17 MeV. Therefore, for the IGrM particles with masses less than 17 MeV the merger becomes problematic. We also define the region in the vicinity of our Local Group where the formation of the Hubble flows starts. For such processes, the zero-acceleration surface (where the gravitational attraction is balanced by the cosmological accelerated expansion) plays the crucial role. We show that such surface is absent for the Local Group. Instead, we find two points and one circle with zero acceleration. Nevertheless, there is a nearly closed area around the MW and M31 where the absolute value of the acceleration is approximately equal to zero. The Hubble flows are formed outside of this area.« less
66 citations
••
03 Mar 2020TL;DR: An automatic deep-learning-based method for stage detection of diabetic retinopathy by single photography of the human fundus and the multistage approach to transfer learning, which makes use of similar datasets with different labeling are proposed.
Abstract: Diabetic retinopathy is one of the most threatening complications of diabetes that leads to permanent blindness if left untreated. One of the essential challenges is early detection, which is very important for treatment success. Unfortunately, the exact identification of the diabetic retinopathy stage is notoriously tricky and requires expert human interpretation of fundus images. Simplification of the detection step is crucial and can help millions of people. Convolutional neural networks (CNN) have been successfully applied in many adjacent subjects, and for diagnosis of diabetic retinopathy itself. However, the high cost of big labeled datasets, as well as inconsistency between different doctors, impede the performance of these methods. In this paper, we propose an automatic deep-learning-based method for stage detection of diabetic retinopathy by single photography of the human fundus. Additionally, we propose the multistage approach to transfer learning, which makes use of similar datasets with different labeling. The presented method can be used as a screening method for early detection of diabetic retinopathy with sensitivity and specificity of 0.99 and is ranked 54 of 2943 competing methods (quadratic weighted kappa score of 0.925466) on APTOS 2019 Blindness Detection Dataset (13000 images).
59 citations
•
TL;DR: In this paper, an automatic deep-learning-based method for stage detection of diabetic retinopathy by single photography of the human fundus was proposed, which makes use of similar datasets with different labels.
Abstract: Diabetic retinopathy is one of the most threatening complications of diabetes that leads to permanent blindness if left untreated. One of the essential challenges is early detection, which is very important for treatment success. Unfortunately, the exact identification of the diabetic retinopathy stage is notoriously tricky and requires expert human interpretation of fundus images. Simplification of the detection step is crucial and can help millions of people. Convolutional neural networks (CNN) have been successfully applied in many adjacent subjects, and for diagnosis of diabetic retinopathy itself. However, the high cost of big labeled datasets, as well as inconsistency between different doctors, impede the performance of these methods. In this paper, we propose an automatic deep-learning-based method for stage detection of diabetic retinopathy by single photography of the human fundus. Additionally, we propose the multistage approach to transfer learning, which makes use of similar datasets with different labeling. The presented method can be used as a screening method for early detection of diabetic retinopathy with sensitivity and specificity of 0.99 and is ranked 54 of 2943 competing methods (quadratic weighted kappa score of 0.925466) on APTOS 2019 Blindness Detection Dataset (13000 images).
57 citations
••
TL;DR: In this article, the influence of tool geometry and process parameters on workpiece delamination and hole quality/integrity was analyzed with respect to workpiece materials, geometrical tool features, and input variables (such as variation in process parameters).
54 citations
Authors
Showing all 821 results
Name | H-index | Papers | Citations |
---|---|---|---|
Andrii Tykhonov | 73 | 270 | 24864 |
Eugene N. Muratov | 41 | 143 | 6550 |
Victor E. Kuz’min | 26 | 129 | 3010 |
Maxim Eingorn | 22 | 108 | 1274 |
Nikolay P. Malomuzh | 18 | 95 | 938 |
Anatoliy G. Goncharuk | 14 | 63 | 511 |
Vitaliy D. Rusov | 12 | 104 | 563 |
Ekaterina V. Varlamova | 10 | 17 | 411 |
Viktor Gogunskii | 9 | 23 | 147 |
Oleksandr Drozd | 8 | 40 | 168 |
T. N. Zelentsova | 8 | 27 | 165 |
Tatyana A. Komleva | 8 | 31 | 176 |
V. P. Smolyar | 8 | 38 | 198 |
Marina Polyakova | 8 | 70 | 201 |
Dmitry A. Maevsky | 7 | 18 | 107 |