Background: Significant progress has been made in the field of machine learning, enabling the development of methods for automatic interpretation of medical images that provide high-quality diagnostics. However, most of these methods require access to confidential data, making them difficult to apply under strict privacy requirements. Existing privacy-preserving approaches, such as federated learning and dataset distillation, have limitations related to data access, visual interpretability, etc. Methods: This study explores the use of generative models to create synthetic medical data that preserves the statistical properties of the original data while ensuring privacy. The research is carried out on the VinDr-Mammo dataset of digital mammography images. A conditional generative method using Latent Diffusion Models (LDMs) is proposed with conditioning on diagnostic labels and lesion information. Diagnostic utility and privacy robustness are assessed via cancer classification tasks and re-identification tasks using Siamese neural networks and membership inference. Results: The generated synthetic data achieved a Fréchet Inception Distance (FID) of 5.8, preserving diagnostic features. A model trained solely on synthetic data achieved comparable performance to one trained on real data (ROC-AUC: 0.77 vs. 0.82). Visual assessments showed that synthetic images are indistinguishable from real ones. Privacy evaluations demonstrated a low re-identification risk (e.g., mAP@R = 0.0051 on the test set), confirming the effectiveness of the privacy-preserving approach. Conclusions: The study demonstrates that privacy-preserving generative models can produce synthetic medical images with sufficient quality for diagnostic task while significantly reducing the risk of patient re-identification. This approach enables secure data sharing and model training in privacy-sensitive domains such as medical imaging.

BACKGROUND: Breast cancer is one of the leading causes of cancer-related mortality in women [1]. Regular mass screening with mammography plays a critical role in the early detection of changes in breast tissue. However, the early stages of pathology often go undetected and are difficult to diagnose [2]. Despite the effectiveness of mammography in reducing breast cancer mortality, manual image analysis can be time consuming and labor intensive. Therefore, attempts to automate this process, for example using computer-aided diagnosis systems, are relevant [3]. In recent years, however, solutions based on neural networks have gained increasing interest, especially in biology and medicine [4-6]. Technological advances using artificial intelligence have already demonstrated their effectiveness in pathology detection [7, 8]. AIM: The study aimed to develop an automated solution to detect breast cancer on mammograms. MATERIALS AND METHODS: The solution is implemented as follows: a deep neural network-based tool has been developed to obtain the probability of malignancy from the input image. A combined dataset from public datasets such as MIAS, CBIS-DDSM, INbreast, CMMD, KAU-BCMD, and VinDr-Mammo [9–14] was used to train the model. RESULTS: The classification model, based on the EfficientNet-B3 architecture, achieved an area under the ROC curve of 0.95, a sensitivity of 0.88, and a specificity of 0.9 when tested on a sample from the combined dataset. The model’s high generalization ability, which is another advantage, was demonstrated by its ability to perform well on images from different datasets with varying data quality and acquisition regions. Furthermore, techniques such as image pre-cropping and augmentations during training were used to enhance the model's performance. CONCLUSIONS: The experimental results demonstrated that the model is capable of accurately detecting malignancies with a high degree of confidence. The obtained high-quality metrics offer a significant potential for implementing this method in automated diagnostics, for instance, as an additional opinion for medical specialists.

Methods and tools for binary code analysis developed in the Institute of System Programming, Russian Academy of Sciences, and their applications in algorithm and data format recovery are considered. The executable code of various general-purpose CPU architectures is analyzed. The analysis is performed given no source codes, debugging information, and specific OS version requirements. The approach implies collecting a detailed machine instruction level execution trace; a method for successively increasing presentation level; extraction of algorithm’s code followed by structuring of both code and data formats it processes. Important results are obtained, viz. an intermediate representation is developed that allows carrying out most preliminary processing tasks and algorithm code extraction without having to focus on specifics of a given machine; and a method and software tool are developed for automated recovery of network message and file formats. The tools are integrated into the unified analysis platform that supports their combined use. The architecture behind the platform is also described. Examples of its application to real programs are given.

В статье представлена система расследования утечек конфиденциальных текстовых документов, где в качестве каналов утечек рассматриваются изображения, полученные путем сканирования/фотографирования напечатанных документов, а также фотографии документов на экране монитора. Таким образом, система нацелена на защиту каналов, не защищаемых традиционными DLP-решениями. В качестве механизма защиты документов выбрано внедрение цифрового водяного знака (ЦВЗ), предполагающее изменение визуального представления документа. В случае анонимной утечки конфиденциального документа ЦВЗ позволит установить сотрудника, допустившего утечку – намеренно или в результате нарушения протокола безопасности. В статье описана архитектура системы, состоящая из клиентской и серверной частей. На рабочие станции сотрудников устанавливаются компоненты, обеспечивающие внедрение ЦВЗ в документы, отправляемые на печать или выводимые на экран монитора. Факты маркирования документов регистрируются и отправляются на удаленный сервер, использующий данную информацию при расследовании утечек аналитиком службы безопасности. Разработаны алгоритмы маркирования для встраивания ЦВЗ в текстовые документы, выводимые на печать и экран монитора. При маркировании документа на экране монитора ЦВЗ встраивается в межстрочные интервалы: цифровая метка кодируется последовательностью затемненных и осветленных областей. Для встраивания ЦВЗ в печатаемые документы разработано три алгоритма маркирования: на основе горизонтального и вертикального смещения слов, а также посредством изменения яркости отдельных фрагментов слов. Разработана методика тестирования алгоритмов маркирования в условиях, приближенным к условиям эксплуатации, оценена область применимости алгоритмов. Проведен анализ вероятных атак на систему, и сформулирована модель нарушителя.

An electrocardiogram (ECG) is one of the most common medical examinations. High-quality interpretation of a 12-channel electrocardiogram is important for subsequent diagnosis and treatment. One of the important steps in deciphering an ECG is to determine the boundaries of the elements of the PQRST complex. The article discusses mathematical methods for determining the boundaries of the P, T waves and the QRS complex, as well as the R, P and T peaks, presents the shortcomings of mathematical methods for determining the elements of the PQRST complex. And also the values ​​of the metrics obtained as a result of training the neural network segmentation model of the PQRST-complex are given. The experiments performed show the relevance of using neural network and combined approaches to the analysis of the PQRST complex.