ASArtjoms Suponenkovs

Artjoms Suponenkovs

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Artjoms Suponenkovs has a total of 8 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Hybrid Approach to Colony-Forming Unit Counting Problem Using Multi-Loss U-Net Reformulation

    by Vilen Jumutc, Artjoms Suponenkovs, Andrey Bondarenko, Dmitrijs Bļizņuks, Alexey Lihachev

    Colony-Forming Unit (CFU) counting is a complex problem without a universal solution in biomedical and food safety domains. A multitude of sophisticated heuristics and segmentation-driven approaches have been proposed by researchers. However, U-Net remains the most frequently cited and used deep learning method in these domains. The latter approach provides a segmentation output map and requires an additional counting procedure to calculate unique segmented regions and detect microbial colonies. However, due to pixel-based targets, it tends to generate irrelevant artifacts or errant pixels, leading to inaccurate and mixed post-processing results. In response to these challenges, this paper proposes a novel hybrid counting approach, incorporating a multi-loss U-Net reformulation and a post-processing Petri dish localization algorithm. Firstly, a unique innovation lies in the multi-loss U-Net reformulation. An additional loss term is introduced in the bottleneck U-Net layer, focusing on the delivery of an auxiliary signal that indicates where to look for distinct CFUs. Secondly, the novel localization algorithm automatically incorporates an agar plate and its bezel into the CFU counting techniques. Finally, the proposition is further enhanced by the integration of a fully automated solution, which comprises a specially designed uniform Petri dish illumination system and a counting web application. The latter application directly receives images from the camera, processes them, and sends the segmentation results to the user. This feature provides an opportunity to correct the CFU counts, offering a feedback loop that contributes to the continued development of the deep learning model. Through extensive experimentation, the authors of this paper have found that all probed multi-loss U-Net architectures incorporated into the proposed hybrid approach consistently outperformed their single-loss counterparts, as well as other comparable models such as self-normalized density maps and YOLOv6, by at least 1% to 3% in mean absolute and symmetric mean absolute percentage errors. Further significant improvements were also reported through the means of the novel localization algorithm. This reaffirms the effectiveness of the proposed hybrid solution in addressing contemporary challenges of precise in vitro CFU counting.


  • Open Access

    Hybrid Approach to Colony-Forming Unit Counting Problem using Multi-Loss U-Net Reformulation

    by Vilen Jumutc, Artjoms Suponenkovs, Andrey Bondarenko, Dmitrijs Bļizņuks, Alexey Lihachev

    The Colony-Forming Unit (CFU) counting problem remains a complex issue without a universal solution in biomedical and food safety domains. A multitude of sophisticated heuristics and segmentation-driven approaches have been proposed by numerous researchers. Among those, U-Net is the most frequently cited and popular Deep Learning method. The latter approach provides a segmentation output map and requires an additional counting procedure which accounts for unique segmented regions and detected microbial colonies. However, because of pixel-based targets it tends to generate irrelevant artifacts or errant pixels, leading to inaccurate and mixed post-processing results. In response to these challenges, we propose a novel hybrid counting approach, incorporating a multi-loss U-Net reformulation and a post-processing Petri dish localization algorithm. First of all, our unique innovation lies in the multi-loss U-Net reformulation. We introduce an additional loss term at the bottleneck U-Net layer, focusing on delivering an auxiliary signal indicating where to look for distinct CFUs. Second, our novel localization algorithm accurately incorporates an agar plate and its bezel into the CFU counting routines. Finally, our proposition is further enhanced by the integration of a fully automated solution. This comprises a specially designed uniform Petri dish illumination system and a counting web application. The latter application is capable of directly receiving images from the camera, which are subsequently processed, and the segmentation results are sent back to the user. This feature provides an opportunity to correct the CFU counts, offering a feedback loop that contributes to the continued development of the Deep Learning model. Through extensive experimentation, we have found that all probed multi-loss U-Net architectures incorporated in our hybrid approach consistently outperform their single-loss counterparts which utilize exclusively the combination of Tversky and Cross-Entropy training losses at the output U-Net layer. We report further significant improvements by the means of our novel localization algorithm. This reaffirms the effectiveness of our proposed hybrid solution in addressing contemporary challenges of the precise in-vitro CFU counting.


  • Development of Computer-Based Diagnostic System in the Domain of Magnetic Resonance Applications

    by Artjoms Supoņenkovs

    The aim of the doctoral thesis is to develop approaches, methods and algorithms that allow to create digital medical diagnostic systems. Medical image pre-processing, segmentation and analysis are the most important aspects of research. This work investigates a problem of pattern recognition, geometric and texture features analysis. Stages of human body fluids and tissues analysis are described in this work. There are many stages of this analysis: medical image pre-processing; fluids and tissues detection and localization; geometric and texture features calculation; defect or lesion detection. Knee joint anatomy, stages of knee osteoarthritis, medical image processing, magnetic resonance imaging are the most important aspects of research and are described in the theoretical part of the thesis. Six new modules have been developed and are described in the practical part. These new modules have a lot of possibilities: relaxation times calculation; human body fluids and tissues segmentation, features calculation and defect detection; medical image pre-processing and visualization; optimization of medical image processing instruction; textual analysis of patient's information. Geometric, texture and topology features of healthy and injured knee joint fluids and tissues have been defined in the experimental part of the doctoral thesis. Pre-processing methods, segmentation methods, author's pattern recognition modes and image process optimization methods have been compared in several experiments. As a result of the doctoral thesis, the new author's MRI medical image processing system has been developed. This system consists of several new modules. These modules can be useful for knee joint diagnostics and allow starting treatment earlier and therefore reducing the risk of cartilage destruction.

  • Datorizētas diagnostikas sistēmas izstrāde magnētiskās rezonanses lietojumsfērā

    by Artjoms Supoņenkovs

    Promocijas darba mērķis ir piedāvāt pieejas, metodes, algoritmus un realizācijas, kas ļauj veikt datorizētu diagnostiku medicīniskā lietojumsfērā. Promocijas darbā izpētītas medicīnisko attēlu pirmapstrādes, segmentācijas un analizēšanas metodes, analizētas šablonu meklēšanas, ģeometrisko īpašību analizēšanas un tekstūras analizēšanas pieejas. Darbā parādīti organiskās struktūras analizēšanas etapi: medicīnisko attēlu pirmapstrāde; medicīnisko attēlu segmentācija; organiskās struktūras noteikšana un lokalizācija; organiskās struktūras fizikālās, ģeometrisko un tekstūru īpašību aprēķināšana; organiskās struktūras bojājumu detektēšana. Realizācijas daļā aprakstīti seši jauni autora izveidoti moduļi. Šiem moduļiem ir vairākas iespējas: relaksācijas laiku noteikšana; audu un šķidrumu segmentācija, analizēšana, lokalizēšana, īpašību noteikšana; medicīnisko attēlu pirmapstrāde un vizualizācija; medicīnisko attēlu apstrādes instrukcijas minimizācija; pacienta anketas analizēšana. Eksperimentālajā daļā noteiktas veselas un bojātas ceļa locītavas audu un šķidrumu raksturīgās fizikālās, tekstūras, ģeometriskās un topoloģiskās pazīmes. Eksperimentos salīdzinātas pirmapstrādes un segmentācijas metodes, autora izstrādāti šablonu meklēšanas režīmi un attēlu apstrādes optimizācijas metodes. Promocijas praktiskā vērtība ir medicīnas magnētiskās rezonanses attēlu apstrādes moduļu komplekss, kas palīdz ārstiem veikt ceļa locītavas neinvazīvu diagnostiku. Darba rezultātu aprobācija notikusi, piedaloties 11 zinātniskās konferencēs, un tā atspoguļota 10 publikācijās starptautiskos zinātniskos krājumos. Darbā ir ievads, piecas nodaļas, secinājumi un 12 pielikumu. Tajā ir 219 lappušu, literatūras sarakstā ir 116 nosaukumu.

  • Open Access

    APPLICATION OF COMPUTER VISION TECHNOLOGIES FOR AUTONOMOUS PILE MANIPULATION

    by Artjoms Suponenkovs, Mihails Kovalovs, Zigurds Markovics

    Modern robots can perform uncreative monotonous tasks. One of such tasks is pile manipulation. Computer vision technologies can help robots acquire additional information by analyzing a pile of complex objects. One of such complex objects is a fish. The presented work investigates the problems of complex object analysis using computer vision. This paper addresses the challenges of image pre-processing, image segmentation, fish detection and occlusion detection. This work results can be useful for developing a computer vision system for pile manipulation.

  • Open Access

    Distributed Indication in LUT-Based Asynchronous Logic

    by Igor Lemberski, Marina Uhanova, Artjoms Suponenkovs