Pavel Mikhailovich Zelter

Pavel Mikhailovich Zelter

Auto-generated profile, based on publicly available information

Pavel Mikhailovich Zelter

Samara State Medical University

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 April 2020 in Computer Optics
Verified
Journal Article
Closed access
SUSamara National Research University
Samara National Research University
YAYu.D. Agafonova
Yu.D. Agafonova
AGA.V. Gaidel
A.V. Gaidel
IRIpsi Ras – Branch Of The Fsrc “Crystallography And Photonics” Ras
Ipsi Ras – Branch Of The Fsrc “Crystallography And Photonics” Ras
PZP.M. Zelter
P.M. Zelter
AKA.V. Kapishnikov
A.V. Kapishnikov
We compare approaches for the automatic detection of pathological changes in brain MRI images that are visible to the naked eye. We analyse multi-stage approaches based on deep learning and threshold processing. A convolutional neural network was formed, a classifier was built based on the use of an ensemble of decision trees, and an algorithm was created for multi-stage image processing. Because of experimental studies, it was found that the most effective method for recognizing images of magnetic resonance imaging is an approach based on an ensemble of decision trees. With its help, 95 % of the images from the test sample were classified correctly. At the same time, using the convolutional neural network, it was possible to classify correctly all images containing the area of pathological changes. The data obtained can be used in practice for the diagnosis of brain diseases, for automating the processing of a large number of studies of magnetic resonance imaging.

28 February 2022 in Cancers
Verified
Journal Article
Open access
PDF file
VEViacheslav Egorov
Viacheslav Egorov
PKPavel Kim
Pavel Kim
AKAlexander Kharazov
Alexander Kharazov
SDSoslan Dzigasov
Soslan Dzigasov
PPPavel Popov
Pavel Popov
ASAlexander Sorokin
Alexander Sorokin
DPCAR’s short- and long-term outcomes are highly diverse, while the causes and prevention of ischemic complications are unclear. To assess oncological, surgical, and hemodynamic outcomes of 40 consecutive DPCARs for pancreatic (n37) and gastric tumors (n3) (2009–2021), retrospective analyses of mortality, morbidity, survival, and hemodynamic consequences after DPCAR were undertaken using case history data, IOUS, and pre- and postoperative CT measurements. In postoperative complications (42.5%), the pancreatic fistula was the most frequent event (27%), 90-day mortality was 7.5. With 27 months median follow-up, median overall (OS) and progression-free survival (PFS) for PDAC were 29 and 18 months, respectively; with 1-, 3-, and 5-years, the OS were 90, 60, and 28%, with an R0-resection rate of 92.5%. Liver and gastric ischemia developed in 0 and 5 (12.5%) cases. Comparison of clinical and vascular geometry data revealed fast adaptation of collateral circulation, insignificant changes in proper hepatic artery diameter, and high risk of ischemic gastropathy if the preoperative diameter of pancreaticoduodenal artery was <2 mm. DP CAR can be performed with acceptable morbidity and survival. OS and RFS in this super-selective cohort were compared to those for resectable cancer. The changes in the postoperative arterial geometry could explain the causes of ischemic complications and determine directions for their prevention.

1 January 2017 in Computer Optics
Verified
Journal Article
Open access
PDF file
SUSamara National Research University
Samara National Research University
НСН. А. Смелкина
Н. А. Смелкина
АКА. В. Колсанов
А. В. Колсанов
SUSamara State Medical University
Samara State Medical University
СЧС. С. Чаплыгин
С. С. Чаплыгин
АХА. Г. Храмов
А. Г. Храмов
We discuss a simple method for automatic recognition of pulmonary emphysema in three-dimensional computer tomography (CT) images. This technique allows one to quantify the disease progress, calculating some numerical characteristics, such as the percentage of the lung tissue affected, as well as visualizing its location and intensity histogram in the region of interest. An experiment on the test data shows that the recognition error is not higher than 7.5%.

1 February 2020 in Computer Optics
Verified
Journal Article
Closed access
SUSamara National Research University
Samara National Research University
TPT.A. Pashina
T.A. Pashina
AGA.V. Gaidel
A.V. Gaidel
IRIpsi Ras – Branch Of The Fsrc “Crystallography And Photonics” Ras
Ipsi Ras – Branch Of The Fsrc “Crystallography And Photonics” Ras
PZP.M. Zelter
P.M. Zelter
Artem NikonorovArtem Nikonorov
Artem Nikonorov
This article discusses the creation of masks for highlighting the lungs in computed tomography images using three methods – the Otsu method, a simple convolutional neural network consisting of 10 identical layers, and the convolutional neural network U-Net. We perform a study and comparison of methods used for automatically highlighting the region of interest (ROI) in computed tomography images of the lungs, which were provided as a courtesy from the Clinics of Samara State Medical University. The solution to this problem is relevant, because medical workers have to manually select the ROI as the first step of the automated processing of lung CT images. An algorithm for post-processing images based on the search for contours, which allows one to improve the quality of segmentation, is proposed. It is concluded that the U-Net highlights the ROI relating to the lung better than the other two methods. At the same time, the simple convolutional neural network highlights the ROI with an accuracy of 97.5%, which is better than the accuracy of 96.7% of the Otsu method and 96.4% of the U-Net.

1 April 2021 in Computer Optics
Verified
Journal Article
Closed access
SUSamara National Research University
Samara National Research University
ASA.A. Sludnova
A.A. Sludnova
VSV.V. Shutko
V.V. Shutko
AGA.V. Gaidel
A.V. Gaidel
PZP.M. Zelter
P.M. Zelter
IRIpsi Ras - Branch Of The Fsrc “Crystallography And Photonics” Ras
Ipsi Ras - Branch Of The Fsrc “Crystallography And Photonics” Ras
This article discusses an idea of a joint analysis of medical images and texts aimed at improving the quality of automated diagnosis of emphysema. We compare the quality of image classification with and without taking into account the localization of the pathology mentioned in radiological reports. The study was carried out on sets of real images of computed tomography of the lungs obtained in clinical studies at Samara State Medical University. It was established that the use of information on the localization of pathology contained in radiological reports leads to an increase in the F-score for the detection from 0.55 to 0.73.

15 March 2021 in I.P. Pavlov Russian Medical Biological Herald
Verified
Journal Article
Closed access
PDF file
OZOlesya V. Zeleva
Olesya V. Zeleva
PZPavel M. Zelter
Pavel M. Zelter
AKAlexander V. Kolsanov
Alexander V. Kolsanov
YPYuliya S. Pyshkina
Yuliya S. Pyshkina
EKEvgeniya K. Kramm
Evgeniya K. Kramm
Detailed knowledge of anatomy and topography of sphenoidal and maxillary sinuses may help to improve surgical approaches, better predict postoperative complications, increase safety of surgical techniques, and, consequently, avoid inflammatory reactions. This also helps to determine the cause of inflammatory changes. Aim. This study aimed to analyze structural types of sphenoidal and maxillary sinuses by examining computed tomography (CT) images of the paranasal sinuses, to determine the frequency of the interposition of the internal carotid artery, and to identify sex- and age-related peculiarities and regularities in the extent of pneumatization and contents in the sphenoidal and maxillary sinuses. Materials and Methods. CT images of 50 patients who were receiving treatment in the otolaryngology unit of the Clinic of SamSMU were analyzed. Toshiba Aquilion 32 CT scanner was used. Results. Patients were divided into three groups according to Hamberger classification of the pneumatization of the sphenoidal sinus by taking into account the topographic relation to the sella turcica using CT data: the presellar type was identified in 10% (n=5) of the cases, the sellar type in 22% (n=11), and the postsellar type in 68% (n=34). The postsellar type is the most common structural type of the sphenoidal sinus. Moreover, patients were divided into groups by the type of pneumatization of the maxillary sinuses. No correlation was found between the pneumatizations of sphenoidal and maxillary sinuses. Moderate negative correlation was observed between age and structural type of the sinuses, and moderate positive correlation was found between the contents of the sinuses. Interposition of the internal carotid arteries was identified in 3 (6%) patients. Conclusions. Our data revealed that the postsellar type is the most common structural type of sphenoidal sinuses. The extent of pneumatization of the sphenoidal and maxillary sinuses is independent of each other. Anatomical peculiarities of the sphenoidal sinus in the form of the interposition of the internal carotid arteries are not very rare and should be taken into account in planning surgical interventions.