Róbert Paulik

Mr. Róbert Paulik

Image Analysis Department, 3DHISTECH Ltd., Budapest, Hungary
orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
14 September 2022 in Sensors
Verified
Journal Article
Open access
PDF file
VJViktor Zoltán Jónás
Viktor Zoltán Jónás
Róbert PaulikRóbert Paulik
Róbert Paulik
MKMiklos Kozlovszky
Miklos Kozlovszky
BMBéla Molnár
Béla Molnár
Ploidy analysis is the fundamental method of measuring DNA content. For decades, the principal way of conducting ploidy analysis was through flow cytometry. A flow cytometer is a specialized tool for analyzing cells in a solution. This is convenient in laboratory environments, but prohibits measurement reproducibility and the complete detachment of sample preparation from data acquisition and analysis, which seems to have become paramount with the constant decrease in the number of pathologists per capita all over the globe. As more open computer-aided systems emerge in medicine, the demand for overcoming these shortcomings, and opening access to even more (and more flexible) options, has also emerged. Image-based analysis systems can provide an alternative to these types of workloads, placing the abovementioned problems in a different light. Flow cytometry data can be used as a reference for calibrating an image-based system. This article aims to show an approach to constructing an image-based solution for ploidy analysis, take measurements for a basic comparison of the data produced by the two methods, and produce a workflow with the ultimate goal of calibrating the image-based system.

17 November 2023 in Sensors
Verified
Journal Article
Open access
PDF file
DKDániel Küttel
Dániel Küttel
LKLászló Kovács
László Kovács
ÁSÁkos Szölgyén
Ákos Szölgyén
Róbert PaulikRóbert Paulik
Róbert Paulik
VJViktor Zoltán Jónás
Viktor Zoltán Jónás
BMBéla Molnár
Béla Molnár
As the field of routine pathology transitions into the digital realm, there is a surging demand for the full automation of microscope scanners, aiming to expedite the process of digitizing tissue samples, and consequently, enhancing the efficiency of case diagnoses. The key to achieving seamless automatic imaging lies in the precise detection and segmentation of tissue sample regions on the glass slides. State-of-the-art approaches for this task lean heavily on deep learning techniques, particularly U-Net convolutional neural networks. However, since samples can be highly diverse and prepared in various ways, it is almost impossible to be fully prepared for and cover every scenario with training data. We propose a data augmentation step that allows artificially modifying the training data by extending some artifact features of the available data to the rest of the dataset. This procedure can be used to generate images that can be considered synthetic. These artifacts could include felt pen markings, speckles of dirt, residual bubbles in covering glue, or stains. The proposed approach achieved a 1–6% improvement for these samples according to the F1 Score metric.

26 October 2021 in Sensors
Verified
Journal Article
Open access
PDF file
Róbert PaulikRóbert Paulik
Róbert Paulik
MKMiklos Kozlovszky
Miklos Kozlovszky
BMBéla Molnár
Béla Molnár
Image quality, resolution and scanning time are critical in digital pathology. In order to create a high-resolution digital image, the scanner systems execute stitching algorithms to the digitized images. Due to the heterogeneity of the tissue sample, complex optical path, non-acceptable sample quality or rapid stage movement, the intensities on pictures can be uneven. The evincible and visible intensity distortions can have negative effect on diagnosis and quantitative analysis. Utilizing the common areas of the neighboring field-of-views, we can estimate compensations to eliminate the inhomogeneities. We implemented and validated five different approaches for compensating output images created with an area scanner system. The proposed methods are based on traditional methods such as adaptive histogram matching, regression-based corrections and state-of-the art methods like the background and shading correction (BaSiC) method. The proposed compensation methods are suitable for both brightfield and fluorescent images, and robust enough against dust, bubbles, and optical aberrations. The proposed methods are able to correct not only the fixed-pattern artefacts but the stochastic uneven illumination along the neighboring or above field-of-views utilizing iterative approaches and multi-focal compensations.

26 September 2024 in Applied Sciences
Verified
Journal Article
Open access
PDF file
VJViktor Zoltán Jónás
Viktor Zoltán Jónás
Róbert PaulikRóbert Paulik
Róbert Paulik
BMBéla Molnár
Béla Molnár
MKMiklos Kozlovszky
Miklos Kozlovszky
Digitization in pathology and cytology labs is now widespread, a significant shift from a decade ago when few doctors used image processing tools. Despite unchanged scanning times due to excitation in fluorescent imaging, advancements in computing power and software have enabled more complex algorithms, yielding better-quality results. This study evaluates three nucleus segmentation algorithms for ploidy analysis using propidium iodide-stained digital WSI slides. Our goal was to improve segmentation accuracy to more closely match DNA histograms obtained via flow cytometry, with the ultimate aim of enhancing the calibration method we proposed in a previous study, which seeks to align image cytometry results with those from flow cytometry. We assessed these algorithms based on raw segmentation performance and DNA histogram similarity, using confusion-matrix-based metrics. Results indicate that modern algorithms perform better, with F1 scores exceeding 0.845, compared to our earlier solution’s 0.807, and produce DNA histograms that more closely resemble those from the reference FCM method.

28 January 2026 in Sensors
Verified
Journal Article
Open access
PDF file
ACAnnamaria Csizmadia
Annamaria Csizmadia
BMBéla Molnár
Béla Molnár
MKMarianna Dimitrova Kucarov
Marianna Dimitrova Kucarov
KKKrisztian Koos
Krisztian Koos
Róbert PaulikRóbert Paulik
Róbert Paulik
Tibor KrenacsTibor Krenacs
Tibor Krenacs
Introduction: Accurate nuclear segmentation is essential for the reliable diagnostic interpretation of fluorescence in situ hybridization (FISH) results. However, automated 2D digital algorithms often fail in samples with dense or overlapping nuclei, such as lymphomas, due to the loss of spatial depth information. Here, we tested if AI-based 3D nuclear segmentation can improve the accuracy, reproducibility, and diagnostic reliability of FISH reading in critical situations. Materials and Methods: Formalin-fixed follicular lymphoma sections were FISH-labeled for BCL2 gene rearrangements and digitally scanned in multilayer Z-stacks. The analytic performance in nuclear segmentation of the adaptive thresholding-based FISHQuant, and the freely accessible AI-based NucleAIzer, StarDist, and Cellpose algorithms, were compared to the eye control-based traditional FISH testing, primarily focusing on nuclear segmentation. Results: We revealed that the Cellpose algorithm showed limited sensitivity to low-intensity signals and the adaptive thresholding 2D segmentation, and FISHQuant struggled to resolve densely packed nuclei, occasionally underestimating their counts. In contrast, 3D segmentation across focal planes significantly improved the nuclear separation and signal localization. AI-driven 3D models, especially NucleAIzer and StarDist, showed improved precision, lower variance (VP/VS ≈ 0.96), and improved gene spot correlation (r > 0.82) across multiple focal planes. The similar average number of gene spots per cell nuclei in the AI-based analyses as the eye control counting, despite the elevated number of cell nuclei found with AI, validated the AI nuclear segmentation results. Conclusions: Inaccurate segmentation limits automated diagnostic FISH signal evaluation. Deep learning 3D approaches, particularly NucleAIzer and StarDist, may overcome thresholding and 2D constraints and improve the consistency of nuclear detection, resulting in better classification of pathogenetic gene aberrations with automated workflows in digital pathology.