Pavel G. Roitberg
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Pavel G. Roitberg has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- A. A. Varfolomeeva
Care Mentor AI, Research and Development Department
1 shared publication - Aleksey I. Sokolov
N.P. Ogarev Mordovian State University
1 shared publication - Deryabina O.N.
National research Mordovia State University named after N. P. Ogarev
1 shared publication - Dmitrii S. Blinov
Dmitry Rogachev National Research Medical Center of Pediatric Hematology, Oncology and Immunology
1 shared publication - E. V. Sushkov
Moscow State Clinical Hospital No. 40, Department of Oncology
1 shared publication - Nadezhda Ivanovna Rozhkova
P.A. Hertzen Moscow Oncology Research Institute – branch of the National Medical Research Radiological Centre
1 shared publication
Latest Publications
- 10 October 2025
- Published by Paediatrician Publishers LLC in Annals of the Russian academy of medical sciences
- Vol. 80 (3) , 198-206
Risks and limitations of using artificial intelligence in medicine
The scientific, technical, ethical, and legal aspects of applying artificial intelligence (AI) in modern medicine, as well as its impact on medical professionals and patients, are examined. The research methodology is based on the analysis of scientific publications dedicated to the use of AI in medicine. Systematization of the available data allows identifying key limitations faced by developers of medical AI systems and their users. These issues are related to the quality and completeness of the data on which machine learning models are built, the limited clinical context, challenges in knowledge generalization, and system interoperability. Ethical challenges include concerns about privacy, algorithmic bias, and the distribution of responsibility. Additionally, there are issues regarding the regulatory framework for AI in healthcare and the training of medical professionals in computer technologies. Overcoming these limitations requires improving data quality, developing multimodal systems, increasing algorithm transparency, and refining the regulatory framework.
- 10 October 2025
- Published by Paediatrician Publishers LLC in Annals of the Russian academy of medical sciences
- Vol. 80 (3) , 198-206
Risks and limitations of using artificial intelligence in medicine
The scientific, technical, ethical, and legal aspects of applying artificial intelligence (AI) in modern medicine, as well as its impact on medical professionals and patients, are examined. The research methodology is based on the analysis of scientific publications dedicated to the use of AI in medicine. Systematization of the available data allows identifying key limitations faced by developers of medical AI systems and their users. These issues are related to the quality and completeness of the data on which machine learning models are built, the limited clinical context, challenges in knowledge generalization, and system interoperability. Ethical challenges include concerns about privacy, algorithmic bias, and the distribution of responsibility. Additionally, there are issues regarding the regulatory framework for AI in healthcare and the training of medical professionals in computer technologies. Overcoming these limitations requires improving data quality, developing multimodal systems, increasing algorithm transparency, and refining the regulatory framework.
- Open Access
- 31 January 2021
- Published by Sechenov University in Sechenov Medical Journal
- Vol. 11 (3) , 4-14
Neural network-based segmentation model for breast cancer X-ray screening
Diagnostic efficiency of breast cancer screening remains one of the most important issues in oncology and radiology. Artificial intelligence technologies are widely used in clinical medicine to effectively solve a number of technological and diagnostic problems. The aim. To develop segmentation neural network model for breast plain radiographs analysis with subsequent study of its clinical effectiveness. Materials and methods. The artificial intelligence-based system was developed to analyze X-ray mammography, аnd included a segmentation neural network with the U-Net architecture, a classification neural architecture ResNet50 with outputting the result using gradient boosting. 15486 X-ray cases were used for training, estimation of diagnostic accuracy and validation of the developed segmental model. All cases were labeled in specially developed software environment LabelCMAITech. The segmentation accuracy was determined by Intersection over Union (IoU) similarity coefficient, the probability of malignancy was calculated using the binary classification metrics. Results. The developed system is represented by a segmentation model based on neural network architecture. The model allows, with high accuracy of 0.8176 and higher, at threshold values on the output neurons of the network of 0.1 and 0.15, to localize X-ray findings that have diagnostic value for determining the likelihood of the presence of breast cancer signs in an X-ray mammographic study — focus, architecture distortion, local asymmetry, calcifications. When comparing the results of machine segmentation and marking of images by a radiologist, it was found that the model is not inferior to the doctor in the accuracy of determining the formations, extra-focal calcifications and intraglandular lymph nodes. Conclusion. The results of this study allow considering the model as an intelligent assistant to a radiologist in the analysis of screening mammographic cases.
- 31 January 2021
- Published by Sechenov University in Sechenov Medical Journal
- Vol. 11 (3) , 4-14
Open AccessNeural network-based segmentation model for breast cancer X-ray screening
Diagnostic efficiency of breast cancer screening remains one of the most important issues in oncology and radiology. Artificial intelligence technologies are widely used in clinical medicine to effectively solve a number of technological and diagnostic problems. The aim. To develop segmentation neural network model for breast plain radiographs analysis with subsequent study of its clinical effectiveness. Materials and methods. The artificial intelligence-based system was developed to analyze X-ray mammography, аnd included a segmentation neural network with the U-Net architecture, a classification neural architecture ResNet50 with outputting the result using gradient boosting. 15486 X-ray cases were used for training, estimation of diagnostic accuracy and validation of the developed segmental model. All cases were labeled in specially developed software environment LabelCMAITech. The segmentation accuracy was determined by Intersection over Union (IoU) similarity coefficient, the probability of malignancy was calculated using the binary classification metrics. Results. The developed system is represented by a segmentation model based on neural network architecture. The model allows, with high accuracy of 0.8176 and higher, at threshold values on the output neurons of the network of 0.1 and 0.15, to localize X-ray findings that have diagnostic value for determining the likelihood of the presence of breast cancer signs in an X-ray mammographic study — focus, architecture distortion, local asymmetry, calcifications. When comparing the results of machine segmentation and marking of images by a radiologist, it was found that the model is not inferior to the doctor in the accuracy of determining the formations, extra-focal calcifications and intraglandular lymph nodes. Conclusion. The results of this study allow considering the model as an intelligent assistant to a radiologist in the analysis of screening mammographic cases.