Vassili Kovalev

• PhD
• Leading Researcher Head of Image Analysis Group at National Academy of Sciences of Belarus

Abstract. Generative diffusion models are a well-established method for generating high-quality images. However, there are studies that show that diffusion models are less privacy-friendly than generative models, such as generative adversarial networks and a growing family of their modifications. The discovered vulnerabilities require in-depth stud...

Segmentation of cell nuclei from three-dimensional (3D) volumetric fluorescence microscopy images is crucial for biological and clinical analyses. In recent years, convolutional neural networks have become the reliable 3D medical image segmentation standard. However, convolutional layers are limited by their finite receptive fields and weight-shari...

Разработан метод обнаружения паттернов реальных пациентов, присутствующих на сгенерированных изображениях компьютерной томографии. Метод включает следующие шаги: корреляция пар сгенерированных и реальных изображений для предварительного отбора пар наиболее близких изображений; вычисление статистик корреляции с использованием прямого и обратного пре...

Интеграция аэрокосмического приборостроения с технологиями искусственного интеллекта требует использования современных нейронных сетей на компактных вычислительных устройствах, таких как мобильные вычислительные устройства, одноплатные компьютеры, ускорители нейронных сетей. В данном исследовании проводится оценка эффективности бюджетного тензорног...

The 2024 ImageCLEFmedical GANs task Controlling the Quality of Synthetic Medical Images created via GANs is in its second edition. It comprises two sub-tasks which address the security and privacy concerns related to personal medical image data in the context of generating and using synthetic images in different real-life scenarios. The first sub-t...

Vertical Federated Learning (VFL) is a privacy-preserving distributed learning paradigm where different parties collaboratively learn models using partitioned features of shared samples, without leaking private data. Recent research has shown promising results addressing various challenges in VFL, highlighting its potential for practical applicatio...

The ImageCLEF evaluation campaign was integrated with CLEF (Conference and Labs of the Evaluation Forum) for more than 20 years and represents a Multimedia Retrieval challenge aimed at evaluating the technologies for annotation, indexing, and retrieval of multimodal data. Thus, it provides information access to large data collections in usage scena...

The 2023 ImageCLEFmedical GANs task is the first edition of this task, examining the existing hypothesis that GANs (Generative Adversarial Networks) are generating medical images that contain the "fingerprints" of the real images used for generative network training. The objective proposed to the participants is to identify the real images that wer...

This paper presents an overview of the ImageCLEF 2023 lab, which was organized in the frame of the Conference and Labs of the Evaluation Forum – CLEF Labs 2023. ImageCLEF is an ongoing evaluation event that started in 2003 and that encourage the evaluation of the technologies for annotation, indexing and retrieval of multimodal data with the goal o...

Objectives . Morphological analysis of papillary thyroid cancer is a cornerstone for further treatment planning. Traditional and neural network methods of extracting parts of images are used to automate the analysis. It is necessary to prepare a set of data for teaching neural networks to develop a system of similar anatomical region in the histopa...

In this paper, we provide an overview of the upcoming ImageCLEF campaign. ImageCLEF is part of the CLEF Conference and Labs of the Evaluation Forum since 2003. ImageCLEF, the Multimedia Retrieval task in CLEF, is an ongoing evaluation initiative that promotes the evaluation of technologies for annotation, indexing, and retrieval of multimodal data...

Нейросетевой программный комплекс для поддержки принятия решений при диагностике заболеваний легких на основе рентгеновских и томографических изображений (далее по тексту – LungExpert) предназначен для автоматизации процессов диагностики онкологических и инфекционных заболеваний легких. LungExpert выполняет компьютерную диагностику компьютерно-томо...

A well-designed CAD tool should respond to input requests, user actions, and perform input checks. Thus, an important element of such a tool is the pre-processing of incoming data and screening out those data that cannot be processed by the application. In this paper, we consider non-trivial methods of chest сomputed tomography (CT) images verifica...

Artificial intelligence (AI) methods for the automatic detection and quantification of COVID-19 lesions in chest computed tomography (CT) might play an important role in the monitoring and management of the disease. We organized an international challenge and competition for the development and comparison of AI algorithms for this task, which we su...

The traditional image descriptor definition algorithms are considered, such as SIFT, ORB, LBP, GLSM. With the help of them, the searching task for a similar anatomical area on the CT images of the lungs is solved. The article proposes a methodology for performing a comparative traditional algorithms for determining images descriptors analysis and o...

This paper presents an overview of the ImageCLEF 2022 lab that was organized as part of the Conference and Labs of the Evaluation Forum – CLEF Labs 2022. ImageCLEF is an ongoing evaluation initiative (first run in 2003) that promotes the evaluation of technologies for annotation, indexing and retrieval of visual data with the aim of providing infor...

ImageCLEF s part of the Conference and Labs of the Evaluation Forum (CLEF) since 2003. CLEF 2022 will take place in Bologna, Italy. ImageCLEF is an ongoing evaluation initiative which promotes the evaluation of technologies for annotation, indexing, and retrieval of visual data with the aim of providing information access to large collections of im...

In this work, we present the results of an experimental study of the problem of image classification under the condition of small image datasets. The performance of both traditional and CNN-based methods is examined and compared based on two benchmark image datasets. The first dataset consisted of 12,000 routine hematoxylin-eosin stained histologic...

Artificial intelligence (AI) has shown promise for diagnosing prostate cancer in biopsies. However, results have been limited to individual studies, lacking validation in multinational settings. Competitions have been shown to be accelerators for medical imaging innovations, but their impact is hindered by lack of reproducibility and independent va...

This work is dedicated to the problem of image classification under the condition of small image datasets. Both traditional and CNN-based methods are examined and compared based on a benchmark image dataset. The dataset consisted of 12000 routine hematoxylin-eosin stained histological images. They represent the biopsy samples of normal tissue and t...

In this paper, we explore the ability of an AI-based computer-aided diagnostic system (CAD) to help to reveal the early signs of probable lung diseases in X-ray images. We use a large screening database which contains natively-digital X-ray images acquired between 2001 and 2014 along with the corresponding diagnostic reports provided by the radiolo...

This paper presents an overview of the ImageCLEF 2021 lab that was organized as part of the Conference and Labs of the Evaluation Forum – CLEF Labs 2021. ImageCLEF is an ongoing evaluation initiative (first run in 2003) that promotes the evaluation of technologies for annotation, indexing and retrieval of visual data with the aim of providing infor...

This paper presents an overview of the ImageCLEF 2021 lab that was organized as part of the Conference and Labs of the Evaluation Forum-CLEF Labs 2021. ImageCLEF is an ongoing evaluation initiative (first run in 2003) that promotes the evaluation of technologies for annotation, indexing and retrieval of visual data with the aim of providing informa...

ImageCLEF is a part of the Conference and Labs of the Evaluation Forum (CLEF) initiative and includes a variety of tasks dedicated to multimodal image information retrieval, including image classification and annotation. The tuberculosis (TB) task is one of the ImageCLEF tasks which started in 2017 and changed from year to year. The 2021 edition wa...

Infrared spectroscopy hampered by physical distortions from scattering and instrumental effects. Therefore, it is generally accepted that spectra should be pre-processed before further data analysis. Deep learning community offers augmentation techniques as an alternative approach to deal with variability in the data. In this paper we propose an Ex...

Artificial intelligence (AI) methods for the automatic detection and quantification of COVID-19 lesions in chest computed tomography (CT) might play an important role in the monitoring and management of the disease. We organized an international challenge and competition for the development and comparison of AI algorithms for this task, which we su...

In this paper, we present three different methods for segmentation of lung lesions associated with COVID-19 virus disease in 3D CT images. All the methods are based on using state-of-the art deep learning techniques. The proposed methods were tested in the framework of COVID-19 Lung CT Lesion Segmentation Challenge 2020 (COVID-19-20), a MICCAI-asso...

This paper presents the ideas for the 2021 ImageCLEF lab that will be organized as part of the Conference and Labs of the Evaluation Forum—CLEF Labs 2021 in Bucharest, Romania. ImageCLEF is an ongoing evaluation initiative (active since 2003) that promotes the evaluation of technologies for annotation, indexing and retrieval of visual data with the...

This paper presents the ideas for the 2021 ImageCLEF lab that will be organized as part of the Conference and Labs of the Evaluation Forum-CLEF Labs 2021 in Bucharest, Romania. ImageCLEF is an ongoing evaluation initiative (active since 2003) that promotes the evaluation of technologies for annotation, indexing and retrieval of visual data with the...

This paper presents the results of an experimental study and the development of tools for automatic analysis and recognition of histological images in order to obtain quantitative estimates of the presence and degree of aggressiveness of prostate cancer in the commonly used Gleason and ISUP scales. The input data consisted of 10 616 whole-slide his...

This study is dedicated to solving the image processing tasks of different types, including detection and analysis of lesions, ыegmentation and recognition of biomedical images with the use of deep learning methods in computerized disease diagnosis. With the use of large datasets consisting of hundreds of thousands of medical images of different mo...

Analysis of breast cancer whole-slide image is an extremely labor-intensive process. Histological whole slide images have the following features: a high degree of tissue diversity both in one image and between different images, hierarchy, a large amount of graphic information and different artifacts. In this work, pre-processing of breast cancer wh...

The research object are biomedical whole slide histological images of breast cancer. The aim of the work is to develop methods, algorithms and basic elements of a software for automatic search of tumor sites, adaptive assessment of the immunohistochemical markers expression and quantitative assessment of the analysis results. At this stage, we have...

ImageCLEF is a part of the Conference and Labs of the Evaluation Forum (CLEF) initiative and presents a set of image information retrieval tasks. ImageCLEF was historically focused on the variety of multimodal image classification, retrieval and annotation tasks. The tuberculosis task started in ImageCLEF in 2017 and changed from year to year. This...

В настоящий момент большинство исследований и разработок в области глубокого обучения концентрируются на повышении точности распознавания, в то время как проблема состязательных атак на глубокие нейронные сети и их последствий пока не получила должного внимания. Данная статья посвящена экспериментальной оценке влияния различных факторов на устойчив...

This paper presents an overview of the ImageCLEF 2020 lab that was organized as part of the Conference and Labs of the Evaluation Forum - CLEF Labs 2020. ImageCLEF is an ongoing evaluation initiative (first run in 2003) that promotes the evaluation of technologies for annotation, indexing and retrieval of visual data with the aim of providing infor...

This paper presents an overview of the ImageCLEF 2020 lab that was organized as part of the Conference and Labs of the Evaluation Forum-CLEF Labs 2020. ImageCLEF is an ongoing evaluation initiative (first run in 2003) that promotes the evaluation of technologies for annotation, indexing and retrieval of visual data with the aim of providing informa...

Приводятся результаты разработки программно-аппаратного комплекса (микромодуля) по обнаружению и классификации изображений подстилающей поверхности Земли. Микромодуль используется на борту легких беспилотных летательных аппаратов (дронов). Полученное устройство имеет размеры 5,2×7,4×3,1 см, массу 52 г, работает на одноплатном микрокомпьютере модели...

A few years ago it was discovered that the deep Convolutional Neural Networks (CNN) are vulnerable to so-called adversarial attacks. An adversarial attack supposes a subtle modification of an original image in such a way that the changes are almost invisible to the human eye. In this work, we are concentrating on biomedical images which are playing...

The purpose of this paper is to present the first version of a statistical atlas of lung lesions. The lesions considered in this study are of different nature and different biological substrates. Despite these lesion types are very common, they may be partly associated with lung tuberculosis because they were detected, isolated, and segmented on lu...

The article presents development results for hardware and software system (micromodule), which detects and classifies underlying surface images of Earth. Given device has size 5.2×7.4×3.1 cm, mass 52 g and uses convolutional neural network based on MobileNetV2 architecture for image classification. The micromodule can be installed on board of a sma...

This paper presents an overview of the 2020 ImageCLEF lab that will be organized as part of the Conference and Labs of the Evaluation Forum—CLEF Labs 2020 in Thessaloniki, Greece. ImageCLEF is an ongoing evaluation initiative (run since 2003) that promotes the evaluation of technologies for annotation, indexing and retrieval of visual data with the...

Fourier‐transform infrared (FTIR) microspectroscopy is rounding the corner to become a label‐free routine method for cancer diagnosis. In order to build infrared‐spectral based classifiers, infrared images need to be registered with Hematoxylin and Eosin (H&E) stained histological images. While FTIR images have a deep spectral domain with thousands...

In this paper, we present detailed results on the success rate of both white-box and black-box untargeted attacks to five types of popular deep Convolutional Neural Network architectures including InceptionV3, Xception, ResNet50, DenseNet121, and Mobilenet. A total of 650,000 images were used for experimentation. They represent chest X-Ray images,...

This paper presents an overview of the 2020 ImageCLEF lab that will be organized as part of the Conference and Labs of the Evaluation Forum-CLEF Labs 2020 in Thessaloniki, Greece. ImageCLEF is an ongoing evaluation initiative (run since 2003) that promotes the evaluation of technologies for annotation, indexing and retrieval of visual data with the...

Breast cancer is one of the most common cancer diseases in the world among women. The reliability of histological verification of breast cancer depends on pathologist's experience, knowledge, his willingness to self-improve and study specialized literature. Digital pathology is also widely used for educational purposes, in telepathology, teleconsul...

In this paper, we study dependence of the success rate of adversarial attacks to the Deep Neural Networks on the biomedical image type, control parameters, and image dataset size. With this work we are going to contribute towards accumulation of experimental results on adversarial attacks for the community dealing with biomedical images. The white-...

In this paper, we explore the possibility of generating artificial biomedical images that can be used as a substitute for real image datasets in applied machine learning tasks. We are focusing on generation of realistic chest X-ray images as well as on the lymph node histology images using the two recent GAN architectures including DCGAN and PGGAN....

Multidrug-resistant tuberculosis (mdrtb) refers to TB infection resistant to at least two most powerful anti-TB drugs, isoniazid and rifampincin. It has been estimated that globally 3.5% (which can be much higher in some regions) of newly diagnosed TB patients, and 20.5% of previously treated patients had mdrtb. Extensively drug-resistant TB (xdrtb...

ImageCLEF is the image retrieval task of the Conference and Labs of the Evaluation Forum (CLEF). ImageCLEF has historically focused on the multimodal and language-independent retrieval of images. Many tasks are related to image classification and the annotation of image data as well as the retrieval of images. Since 2017, when the tuberculosis task...

This paper presents an overview of the ImageCLEF 2019 lab, organized as part of the Conference and Labs of the Evaluation Forum - CLEF Labs 2019. ImageCLEF is an ongoing evaluation initiative (started in 2003) that promotes the evaluation of technologies for annotation, indexing and retrieval of visual data with the aim of providing information acc...

The article describes methods and algorithms related to the analysis of dynamically changing discrete random fields. Time-optimal strategies for the localization of pulsed-point sources having a random spatial distribution and indicating themselves by generating instant delta pulses at random times are proposed. An optimal strategy is a procedure t...

Данная работа посвящена исследованию обнаруженной несколько лет назад проблемы уязвимости глубоких нейронных сетей к так называемым состязательным атакам (Adversarial Attacks), которые заставляют сеть принимать ошибочные классификационные реше-ния. Состязательные атаки осуществляются с помощью «атакующих изображений», которые яв-ляются незначительн...

In this paper we explore the problem of a preliminary computerized diagnosis of lung pathologies under condition of X-ray chest screening.

Image segmentation is a widely used technique to select a region of interest for further work. To assess segmentation impact on the accuracy of classification we used lungs segmentation in the classification of chest X-Ray images by subjects' age groups. Testing of different combinations of segmentation and classification methods showed that the us...

This paper presents an overview of the foreseen ImageCLEF 2019 lab that will be organized as part of the Conference and Labs of the Evaluation Forum - CLEF Labs 2019. ImageCLEF is an ongoing evaluation initiative (started in 2003) that promotes the evaluation of technologies for annotation, indexing and retrieval of visual data with the aim of prov...

In this paper, we study the dependence of the success rate of adversarial attacks to the Deep Neural Networks on the biomedical image type, control parameters, and image dataset size. With this work, we are going to contribute to-wards accumulation of experimental results on adversarial attacks for the community dealing with biomedical images. The...

In this paper, we explore the possibility of generating artificial bio-medical images that can be used as a substitute for real image datasets in applied machine learning tasks. We are focusing on the generation of realistic chest x-ray images as well as the lymph node histology images using the two recent GAN architectures including DCGAN and PGGA...

This paper presents an overview of the foreseen ImageCLEF 2019 lab that will be organized as part of the Conference and Labs of the Evaluation Forum-CLEF Labs 2019. ImageCLEF is an ongoing evaluation initiative (started in 2003) that promotes the evaluation of technologies for annotation, indexing and retrieval of visual data with the aim of provid...

Аннотация. Рассматривается проблема обнаружения и сегментации опухолей на полнослайдовых гисто-логических изображениях с целью компьютерной поддержки процессов диагностики рака молочной же-лезы. В качестве базовых инструментов используются технология глубокого обучения и классификаци-онные сверточные нейронные сети. Основу используемого метода сегм...

ImageCLEF is the image retrieval task of the Conference and Labs of the Evaluation Forum (CLEF). ImageCLEF has historically focused on the multimodal and language-independent retrieval of images. Many tasks are related to image classification and the annotation of image data as well as the retrieval of images. The tuberculosis task was held for the...

The paper presents image description and classification method which was used by United Institute of Informatics Problems (UIIP_BioMed) group for accomplishing the three subtasks of ImageCLEFtuberculosis task. The image description method employed is based on automated detection of tuberculosis (TB) lesions of different types in 3D lung Computed T...

Tumor proliferation is an important biomarker indicative of the prognosis of breast cancer patients. Assessment of tumor proliferation in a clinical setting is highly subjective and labor-intensive task. Previous efforts to automate tumor proliferation assessment by image analysis only focused on mitosis detection in predefined tumor regions. Howev...

Accurate scoring of the severity of pulmonary tuberculosis (TB) is an important problem of quantitative assessment of patients' state. A number of studies were undertaken in order to introduce a scoring system which provide high diagnostic accuracy and reproducibility of TB diagnosis based on chest radiographs and some additional data. The purpose...

Данная работа посвящена задаче генерации правдоподобных (т.е. трудноотличимых от реальных) рентгеновских изображений грудной клетки человека в норме. Указанная задача решается с использованием генеративно-состязательных нейронных сетей (Generative Adversarial Nets). Степень правдоподобия получаемых результатов оценивается как визуально, так и колич...

Данная работа посвящена вопросу анализа базового аппаратного и программного обеспечения существующих недорогих, коммерчески-доступных вычислительных микроплатформ с целью выбора подходящего решения при разработке бортового микромодуля предварительного распознавания и отбора изображений подстилающих поверхностей заданных типов. Предполагается, что с...

The TB Portals Program is an international consortium of physicians, radiologists, and microbiologists from countries with a heavy burden of drug-resistant tuberculosis working with data scientists and IT professionals. Together, we have built the TB Portals, a repository of socioeconomic/geographic, clinical, laboratory, radiological, and genomic...

The paper presents image description and classification methods which were used by United Institute of Informatics Problems (UIIP) group for tuberculosis image classification task. A method based on co-occurrence of adjacent supervoxels in 3D computed tomography (CT) images was used for subtask #1 which was dedicated to image-based recognition of m...

This paper presents an overview of the ImageCLEF 2017 evaluation campaign, an event that was organized as part of the CLEF (Conference and Labs of the Evaluation Forum) labs 2017. ImageCLEF is an ongoing initiative (started in 2003) that promotes the evaluation of technologies for annotation, indexing and retrieval for providing information access...

Recently, due to a number of demonstrated top-ranked achievements the deep learning methods and convolutional neural networks (CNNs) are of high interest for medical image analysis community. The purpose of this study is to compare abilities of CNNs and conventional methods on the specific benchmarking task of image classification and prediction of...

Purpose Automatic detection of lung lesions is a complicated problem due to a large variety of lesion types. Lung lesions could be very different in size (e.g., nodules and lung masses). They may have different location and different internal structure. For instance, the internal structure of lung cancer tumors looks like a solid neoplasm whereas l...

According to the World Health Organization (WHO) lung cancer remains the leading cause of death of men among all malignant tumors [1, 2]. One of the reasons of such a statistics is the fact that the lung cancer is hardly diagnosed on the yearly stages when it is almost asymptomatic. The purpose of this paper is to present a Computer-Aided Diagnosis...

This paper presents results that were obtained in comparative study of the efficiency of conventional and Deep Learning methods on the problem of predicting subjects' age by their chest radiographs. A large study group consisting of chest radiographs of 10 000 people was created by random sub-sampling of suitable subjects from the input image repos...

This paper presents a generalized approach for computing image gradient. It is predominantly aimed at detecting unclear and in certain circumstances even completely invisible borders in large 2D and 3D texture images. The method exploits the conventional approach of sliding window. Once two pixel/voxel sets are sub-sampled from orthogonal window ha...

Everyone wants to automate routine or tiring work. There are a lot of tasks that may be done by drones. For example border protection or delivering. But before any company or even country adopts any technology we need to verify that it's not vulnerable to any attacks. A satellite navigation is at least one vulnerability for drones, which can be eas...

This paper presents results of the first, exploratory stage of research and developments on segmentation of lungs in X-Ray chest images (Chest Radiographs) using Deep Learning methods and Encoder-Decoder Convolutional Neural Networks (ED-CNN). Computational experiments were conducted using GPU Nvidia TITAN X equipped with 3072 CUDA Cores and 12Gb o...

This paper present results of the use of Deep Learning approach and Convolutional Neural Networks (CNN) for the problem of breast cancer diagnosis. Specifically, the main goal of this particular study was to detect and to segment (i.e. delineate) regions of micro-and macro-metastases in whole-slide images of lymph node sections. The whole-slide ima...