JPJarmila Pavlovicova

Jarmila Pavlovicova

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Jarmila Pavlovicova has a total of 41 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Annotation-Free Vision–Language Concept Bottleneck Models for Interpretable Medical Image Classification: A Review

    by Tomas Tanczos, Jarmila Pavlovicova, Veronika Kurilova, Wanda Benesova

    Concept Bottleneck Models (CBMs) route predictions through a layer of human-understandable concepts, but classical CBMs need a concept label for every training image, a cost that is high in medicine. Vision–language models (VLMs) remove that requirement because, in a shared image–text embedding space, a large language model can propose candidate concepts and contrastive similarity can score them, with no concept labels. This review surveys annotation-free VLM-based CBMs and their move into clinical imaging. We cover vision–language foundations (CLIP, ALIGN, LLaVA) and trace concept models from early supervised methods (CBM, CEM, TCAV) to versions that work without manual labels (LaBo, Label-Free CBM, PCBM, OpenCBM, VLG-CBM). We then map these foundations to specific medical fields such as radiology, pathology, cardiology, dermatology, and ophthalmology, highlighting the medical VLM-CBMs closest to actual deployment (CBVLM, AdaCBM, ConcepPath, CBM-RAG). We focus on information leakage, in which task information reaches the prediction through the concept layer without being carried by the named concepts. Two of the seven surveyed medical VLM-CBMs test whether their predictions depend on individual concepts, but none measures leakage, and no shared protocol exists. We identify this gap as a central challenge for interpretable medical AI.

  • Open Access

    Robot-Centric Elevation Map Completion with Sensor Geometry-Aware Augmentation and Uncertainty Estimation

    by Jozef Goga, Michal Kovac, Martin Dekan, Jarmila Pavlovicova, Frantisek Duchon

    Robot-centric elevation maps built from onboard sensing are always incomplete: occlusions, a limited field of view, and range limits leave large unobserved regions that traversability analysis and motion planning must still reason about. We present a supervised framework that completes these maps and reports a per-cell uncertainty. Its core is a ray-cone augmentation that removes angular sectors anchored at the sensor origin during training; unlike the random masks of image inpainting, these sectors match the coverage gaps of real deployments, such as camera failures or reduced camera configurations. Partial maps generated from four depth cameras along legged-robot trajectories in the TartanGround dataset are paired with dense ground truth, yielding 32,329 samples across five outdoor environments. An encoder–decoder network is trained with a masked β-NLL loss and evaluated with a five-fold leave-one-environment-out protocol. The augmentation lowers the completion error on missing sensor sectors by 8.3 to 9.7%, depending on the sector width, at no measurable cost on uncorrupted partial inputs. The completed maps reach a hole root-mean-square error of 2.86 m, a 45% improvement over the strongest classical interpolation baseline.

  • Open Access

    Unconditional GAN-Based Generation of Stage-Specific Fundus Images for Augmenting Diabetic Retinopathy Screening

    by Michal Kovac, Peter Macsik, Veronika Kurilova, Ondrej Straka, Zlatica Fellner, Jarmila Pavlovicova

    Class imbalance in diabetic retinopathy (DR) datasets, particularly the underrepresentation of advanced stages such as proliferative DR, limits classifier performance and necessitates specialized augmentation strategies. This study presents a data augmentation approach using StyleGAN2-ADAto generate stage-specific synthetic fundus images. Unlike conditional approaches requiring labels or segmentation masks, our method trains separate unconditional models for each DR severity stage, producing anatomically realistic images while preserving pathological characteristics. We propose a comprehensive evaluation framework comprising three components: quantitative assessment using distribution-based metrics (FID, KID, IS) and nearest-neighbor-based metrics (NN-SSIM, NN-PSNR); task-based evaluation through DR classification experiments with ConvNeXt Tiny, MobileNet V3 Small, and Xception architectures; and expert validation by three ophthalmologists assessing clinical plausibility. Classification experiments demonstrated improvements in overall accuracy, with ConvNeXt Tiny achieving 73.48% accuracy (+1.82% over baseline) when combining synthetic and classical augmentation. Inter-rater reliability analysis (κ = 0.723 between two evaluators) confirmed the clinical realism of generated images. Our unconditional StyleGAN2-ADA approach provides a scalable, mask-free augmentation solution addressing class imbalance in fundus imaging datasets, potentially enhancing DR screening systems in resource-constrained clinical settings.


  • Open Access

    Review of Smartphone Deep Learning Applications Using Eye Imaging Diagnostic Techniques in Ophthalmology

    by V. Kurilova, J. Goga, A. Thurzo, J. Pavlovicova, M. Oravec, P. Kolar, N. Majtanova

    Eye diseases, which often lead to severe vision loss, can be prevented, reversed, or slowed down, if diagnosed early. The demand for accurate and timely diagnosis has stimulated the exploration of innovative diagnostic techniques, including the connection of deep learning with smartphone applications to analyze eye images. This review explores how smartphone-based systems transform diagnostic processes using built-in cameras, computational resources, and deep learning models to provide inexpensive solutions for early detection on a massive scale.We provide a comprehensive analysis of recent advances in smartphone deep learning applications using eye imaging diagnostic techniques in ophthalmology, highlighting the key application areas and examining the underlying methodologies, hardware, and software solutions used, as well as their clinical implications.We describe 26 complete diagnostic solutions consisting of the smartphone application with the ability to capture images using the smartphone and the accessible deep learning model, supplemented with 48 partial solutions. The aim of this review is to present an overview of current solutions and provide a comprehensive view of future directions for technicians and ophthalmologists in this emerging interdisciplinary research field.


  • Open Access

    Personalization of Optimal Chemotherapy Dosing Based on Estimation of Uncertain Model Parameters Using Artificial Neural Network

    by Martin Dodek, Zuzana Vitková, Anton Vitko, Jarmila Pavlovičová, Eva Miklovičová

    Background/Objectives: The effectiveness of chemotherapy in cancer treatment is often compromised by inter-patient variability, leading to suboptimal outcomes. Traditional dosing protocols rely on population-based models that do not account for individual patient responses and the cancer phenotype. This study aims to develop a personalized chemotherapy dosing strategy by estimating uncertain model parameters using artificial neural networks, ensuring an optimal and individualized treatment approach. Methods: A dynamical model of tumor growth, immune response, and chemotherapy effects is used as the foundation for personalization. A training dataset is generated by simulating state responses across a diverse population of virtual patients, capturing inter-subject variability. The state responses are parameterized (approximated) using the sum of exponential functions to reduce dimensionality, and a multilayer perceptron artificial neural network is trained to estimate patient-specific model parameters based on response data from a single chemotherapy dose. Results: The proposed method effectively estimates patient-specific model parameters, significantly reducing uncertainty compared to conventional population-based models or the nonlinear least squares method. Numerical experiments demonstrate that personalized chemotherapy dosing, optimized using the estimated model parameters, achieves fast tumor remission while minimizing the total drug amount. Conclusions: By integrating the artificial neural network as the parameter estimator with model-based optimization, this study presents a novel approach to personalized chemotherapy dosing.


  • Open Access

    Robust Control of Repeated Drug Administration with Variable Doses Based on Uncertain Mathematical Model

    by Zuzana Vitková, Martin Dodek, Eva Miklovičová, Jarmila Pavlovičová, Andrej Babinec, Anton Vitko

    The aim of this paper was to design a repeated drug administration strategy to reach and maintain the requested drug concentration in the body. Conservative designs require an exact knowledge of pharmacokinetic parameters, which is considered an unrealistic demand. The problem is usually resolved using the trial-and-error open-loop approach; yet, this can be considered insufficient due to the parametric uncertainties as the dosing strategy may induce an undesired behavior of the drug concentrations. Therefore, the presented approach is rather based on the paradigms of system and control theory. An algorithm was designed that computes the required doses to be administered based on the blood samples. Since repeated drug dosing is essentially a discrete time process, the entire design considers the discrete time domain. We have also presented the idea of applying this methodology for the stabilization of an unstable model, for instance, a model of tumor growth. The simulation experiments demonstrated that all variants of the proposed control algorithm can reach and maintain the desired drug concentration robustly, i.e., despite the presence of parametric uncertainties, in a way that is superior to that of the traditional open-loop approach. It was shown that the closed-loop control with the integral controller and stabilizing state feedback is robust against large parametric uncertainties.


  • Open Access

    Structure and Base Analysis of Receptive Field Neural Networks in a Character Recognition Task

    by Jozef Goga, Radoslav Vargic, Jarmila Pavlovicova, Slavomir Kajan, Milos Oravec

    This paper explores extensions and restrictions of shallow convolutional neural networks with fixed kernels trained with a limited number of training samples. We extend the work recently done in research on Receptive Field Neural Networks (RFNN) and show their behaviour using different bases and step-by-step changes within the network architecture. To ensure the reproducibility of the results, we simplified the baseline RFNN architecture to a single-layer CNN network and introduced a deterministic methodology for RFNN training and evaluation. This methodology enabled us to evaluate the significance of changes using the (recently widely used in neural networks) Bayesian comparison. The results indicate that a change in the base may have less of an effect on the results than re-training using another seed. We show that the simplified network with tested bases has similar performance to the chosen baseline RFNN architecture. The data also show the positive impact of energy normalization of used filters, which improves the classification accuracy, even when using randomly initialized filters.


  • Open Access

    Frequency, Time, Representation and Modeling Aspects for Major Speech and Audio Processing Applications

    by Juraj Kacur, Boris Puterka, Jarmila Pavlovicova, Milos Oravec

    There are many speech and audio processing applications and their number is growing. They may cover a wide range of tasks, each having different requirements on the processed speech or audio signals and, therefore, indirectly, on the audio sensors as well. This article reports on tests and evaluation of the effect of basic physical properties of speech and audio signals on the recognition accuracy of major speech/audio processing applications, i.e., speech recognition, speaker recognition, speech emotion recognition, and audio event recognition. A particular focus is on frequency ranges, time intervals, a precision of representation (quantization), and complexities of models suitable for each class of applications. Using domain-specific datasets, eligible feature extraction methods and complex neural network models, it was possible to test and evaluate the effect of basic speech and audio signal properties on the achieved accuracies for each group of applications. The tests confirmed that the basic parameters do affect the overall performance and, moreover, this effect is domain-dependent. Therefore, accurate knowledge of the extent of these effects can be valuable for system designers when selecting appropriate hardware, sensors, architecture, and software for a particular application, especially in the case of limited resources.


  • Open Access

    Using a State-Bounding Observer to Predict the Guaranteed Limits of Drug Amounts in Rats after Oral Administration Based on an Uncertain Pharmacokinetic Model

    by Zuzana Vitková, Martin Dodek, Jarmila Pavlovičová, Anton Vitko

    In the first part of this paper, the problem of using an uncertain pharmacokinetic model is resolved to determine drug concentrations in rats after the oral administration of drug suspensions with and without added tenside. To this end, a generalized pharmacokinetic model determining the guaranteed limits of drug concentrations was designed. Based on this, the design of the so-called state-bounding observer is described in the second part. Rather than being driven by the output of the pharmacokinetic model, the observer can be driven exclusively by a concentration collected from a suitable part of the body and predict the possible risk of the drug concentration not remaining within the therapeutic range for a sufficiently long time. Specifically, the observer determines the upper and lower limits of the concentrations in all the compartments, especially those that are inaccessible for the collection of samples. The proposed approaches are demonstrated by examples.