Radim Kolář | Brno University of Technology (original) (raw)

Papers by Radim Kolář

Lecture notes in electrical engineering, Dec 31, 2022

Pregnancy Hypertension, Feb 29, 2024

This dataset contains microscopic images of multiple cell lines captured by multiple microcopic w... more This dataset contains microscopic images of multiple cell lines captured by multiple microcopic without use of any fluorescent labeling and a manually annotated ground truth for subsequent use in segmentation algorithms. Dataset also includes images reconstructed according to the methods described below in order to ease further segmentation. Our data consist of 244 labelled images of PC-3 (7,907 cells), 205 labelled PNT1A (9,288 cells), in the paper designated as "QPI_Seg_PNT1A_PC3", and 1,819 unlabelled images with a mixture of 22Rv1, A2058, A2780, A8780, DU145, Fadu, G361, HOB and LNCaP used for pretraining, in the paper designated as "QPI_Cell_unlabelled". See Vicar et al. XXXX 2021 DOI XXX (TBA after publishing) Code using this dataset is available at XXXX (TBA after publishing) Materials and methods A set of adherent cell lines of various origins, tumorigenic potential, and morphology we...

Dynamic contrast enhanced ultrasound (DCE-US) imaging is a promising modality for noninvasive ass... more Dynamic contrast enhanced ultrasound (DCE-US) imaging is a promising modality for noninvasive assessment of various tissue kinetic parameters (e.g. time-to-peak, mean transit time (MTT), blood flow etc.). These parameters can be used for diagnosis of pathological tissue functions caused by different diseases. Fast bolus method is based on the fast injection of bolus of contrast agent and repeated imaging with low MI to track contrast agent in tissue of interest (TOI) in time. Mathematical model fitting is an important part of this analysis, which usually presumes that injected bolus is infinitely short, but in practice this is impossible. The main aim of this paper is an analysis of influence of bolus injection rate on MTT estimated by different fitting model. We discover that using a finite-width bolus causes systematic error in MTT estimation.

2020 Computing in Cardiology Conference (CinC)

ECG is a biological signal specific for each person that is hard to create artificially. Therefor... more ECG is a biological signal specific for each person that is hard to create artificially. Therefore, its usage in biometry is highly investigated. It may be assumed that in the future, ECG for biometric purposes will be measured by wearable devices. Therefore, the quality of the acquired data will be worse compared to ambulatory ECG. In this study, we proposed and tested three different ECG-based authentication methods on data measured by Maxim Integrated wristband. Specifically, 29 participants were involved. The first method extracted 22 time-domain featuresintervals and amplitudes from each heartbeat and Hjorth descriptors of an average heartbeat. The second method used 320 features extracted from the wavelet domain. For both methods a random forest was used as a classifier. The deep learning method was selected as the third method. Specifically, the 1D convolutional neural network with embedded feed-forward neural network was used to classify the raw signal of every heartbeat. The first method reached an average false acceptance rate (FAR) 7.11% and false rejection rate (FRR) 6.49%. The second method reached FAR 6.96% and FRR 21.61%. The third method reached FAR 0.57% and FRR 0.00%.

Biophysical Journal

Cell viscoelastic properties are affected by the cell cycle, differentiation, pathological proces... more Cell viscoelastic properties are affected by the cell cycle, differentiation, pathological processes such as malignant transformation. Therefore, evaluation of the mechanical properties of the cells proved to be an approach to obtaining information on the functional state of the cells. Most of the currently used methods for cell mechanophenotypisation are limited by low robustness or the need for highly expert operation. In this paper, the system and method for viscoelasticity measurement using shear stress induction by fluid flow is described and tested. Quantitative Phase Imaging (QPI) is used for image acquisition because this technique enables to quantify optical path length delays introduced by the sample, thus providing a label-free objective measure of morphology and dynamics. Viscosity and elasticity determination were refined using a new approach based on the linear system model and parametric deconvolution. The proposed method allows high-throughput measurements during live cell experiments and even through a time-lapse, where we demonstrated the possibility of simultaneous extraction of shear modulus, viscosity, cell morphology, and QPI-derived cell parameters like circularity or cell mass. Additionally, the proposed method provides a simple approach to measure cell refractive index with the same setup, which is required for reliable cell height measurement with QPI, an essential parameter for viscoelasticity calculation. Reliability of the proposed viscoelasticity measurement system was tested in several experiments including cell types of different Young/shear modulus and treatment with cytochalasin D or docetaxel, and an agreement with atomic force microscopy was observed. The applicability of the proposed approach was also confirmed by a time-lapse experiment with cytochalasin D washout, where an increase of stiffness corresponded to actin repolymerisation in time.

2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)

In this contribution, we focused on optimising a dynamic flow-based shear stress system to achiev... more In this contribution, we focused on optimising a dynamic flow-based shear stress system to achieve a reliable platform for cell shear modulus (stiffness) and viscosity assessment using quantitative phase imaging. The estimation of cell viscoelastic properties is influenced by distortion of the shear stress waveform, which is caused by the properties of the flow system components (i.e., syringe, flow chamber and tubing). We observed that these components have a significant influence on the measured cell viscoelastic characteristics. To suppress this effect, we applied a correction method utilizing parametric deconvolution of the flow system's optimized impulse response. Achieved results were compared with the direct fitting of the Kelvin-Voigt viscoelastic model and the basic steady-state model. The results showed that our novel parametric deconvolution approach is more robust and provides a more reliable estimation of viscosity with respect to changes in the syringe's compliance compared to Kelvin-Voigt model.

Publication in the conference proceedings of EUSIPCO, Florence, Italy, 2006

Publication in the conference proceedings of EUSIPCO, Barcelona, Spain, 2011

IFMBE Proceedings, 2018

Fluorescence microscopy is a golden standard for contemporary biological studies. However, since ... more Fluorescence microscopy is a golden standard for contemporary biological studies. However, since fluorescent dyes cross-react with biological processes, a label-free approach is more desirable. The aim of this study is to create artificial, fluorescence-like nuclei labeling from label-free images using Convolution Neural Network (CNN), where training data are easy to obtain if simultaneous label-free and fluorescence acquisition is available. This approach was tested on holographic microscopic image set of prostate non-tumor tissue (PNT1A) and metastatic tumor tissue (DU145) cells. SegNet and U-Net were tested and provide "synthetic" fluorescence staining, which are qualitatively sufficient for further analysis. Improvement was achieved with addition of bright-field image (by-product of holographic quantitative phase imaging) into analysis and two step learning approach, without and with augmentation, were introduced. Reconstructed staining was used for nucleus segmentation where 0.784 and 0.781 dice coefficient (for DU145 and PNT1A) were achieved. Keywords Deep learning Á Quantitative phase imaging Cell analysis Á Cell nuclei segmentation 2 Materials and Methods 2.1 Experimental Data Adherent cell lines from prostate non-tumor tissue (PNT1A) and metastatic tumor tissues (DU145) were used in the experiment. All cell lines were cultured in RPMI-1640

2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), 2021

This paper focuses on the analysis of image sequences acquired during fast photobleaching using a... more This paper focuses on the analysis of image sequences acquired during fast photobleaching using a standard wide-field microscope. We show that the photobleaching rate estimated for each pixel is not constant for the whole field of view, but it provides a new spatially variant parametric image related to the cell structure and diffusion of fluorophores. We also provide an alternative way to estimate a pixel-wise photobleaching rate with significantly less computation time than exponential model fitting. Clinical Relevance-This method provides an additional way how fluorescence photobleaching might be used for increasing the image contrast

Graefe's Archive for Clinical and Experimental Ophthalmology, 2020

Purpose To measure parameters of the cardiac cycle-induced pulsatile light absorption signal (ple... more Purpose To measure parameters of the cardiac cycle-induced pulsatile light absorption signal (plethysmography signal) of the optic nerve head (ONH) and to compare parameters between normal subjects and patients with different stages of glaucoma. Patients and methods A recently developed video ophthalmoscope was used to acquire short video sequences (10 s) of the ONH. After image registration and trend correction, the pulsatile changing light absorption at the ONH tissue (excluding large vessels) was calculated. The changing light absorption depends on the pulsatile changing blood volume. Various parameters, including peak amplitude, steepness, time-to-peak, full width at half maximum (FWHM), and pulse duration, were calculated for averaged individual pulses (heartbeats) of the plethysmography signal. This method was applied to 19 healthy control subjects and 91 subjects with ocular hypertension, as well as different stages of primary open-angle glaucoma (17 subjects with ocular hype...

Physiological Research, 2010

It has been shown that besides positron emission tomography, single photon emission computed tomo... more It has been shown that besides positron emission tomography, single photon emission computed tomography and magnetic resonance imaging; contrast echocardiography can be used for qualitative and quantitative myocardial perfusion assessment. In this review, the properties of ultrasound contrast agents, imaging techniques and acquisition methods are shortly described and the possibilities of perfusion echocardiography are summarized. The main focus is put on the description of three perfusion models: mathematical models, physical models assuming an ideal inflow and physical models including inflow measurement.

BMC Bioinformatics, 2019

Background: Because of its non-destructive nature, label-free imaging is an important strategy fo... more Background: Because of its non-destructive nature, label-free imaging is an important strategy for studying biological processes. However, routine microscopic techniques like phase contrast or DIC suffer from shadow-cast artifacts making automatic segmentation challenging. The aim of this study was to compare the segmentation efficacy of published steps of segmentation work-flow (image reconstruction, foreground segmentation, cell detection (seed-point extraction) and cell (instance) segmentation) on a dataset of the same cells from multiple contrast microscopic modalities. Results: We built a collection of routines aimed at image segmentation of viable adherent cells grown on the culture dish acquired by phase contrast, differential interference contrast, Hoffman modulation contrast and quantitative phase imaging, and we performed a comprehensive comparison of available segmentation methods applicable for label-free data. We demonstrated that it is crucial to perform the image reconstruction step, enabling the use of segmentation methods originally not applicable on label-free images. Further we compared foreground segmentation methods (thresholding, feature-extraction, level-set, graph-cut, learning-based), seed-point extraction methods (Laplacian of Gaussians, radial symmetry and distance transform, iterative radial voting, maximally stable extremal region and learning-based) and single cell segmentation methods. We validated suitable set of methods for each microscopy modality and published them online. Conclusions: We demonstrate that image reconstruction step allows the use of segmentation methods not originally intended for label-free imaging. In addition to the comprehensive comparison of methods, raw and reconstructed annotated data and Matlab codes are provided.

Biomedical Optics Express, 2018

We describe a low-cost, easy to use binocular instrument to acquire retinal video sequences of bo... more We describe a low-cost, easy to use binocular instrument to acquire retinal video sequences of both eyes simultaneously. After image registration, cardiac cycle-induced pulsatile light attenuation changes can be measured quantitatively with high spatial and temporal resolution. Parameters such as amplitude, pulse form, and time shift between light attenuation changes can be calculated and compared between eye sides. Deviation from intereye symmetry can be not only an early sign of beginning eye diseases such as glaucoma but also a sign of pathological changes in the carotid arteries; hence, this method can improve the early detection of pathological changes. Important features compared to existing monocular instruments are a narrow band light source with the wavelength close to the peak of the blood extinction, and a proportional relationship of image intensity and light intensity, which are the main requirements for quantitative evaluation.

Radioengineering, 2004

This Ω estimator is well accepted, but there are many estimators for the second parameter m, whic... more This Ω estimator is well accepted, but there are many estimators for the second parameter m, which determines the shape of the PDF function. An estimate (m~) of this parameter can be evaluated through the Maximum Likelihood (ML) method. It has been shown that this estimate leads to [11]

Biomedical engineering online, 2016

Analysis of fast temporal changes on retinas has become an important part of diagnostic video-oph... more Analysis of fast temporal changes on retinas has become an important part of diagnostic video-ophthalmology. It enables investigation of the hemodynamic processes in retinal tissue, e.g. blood-vessel diameter changes as a result of blood-pressure variation, spontaneous venous pulsation influenced by intracranial-intraocular pressure difference, blood-volume changes as a result of changes in light reflection from retinal tissue, and blood flow using laser speckle contrast imaging. For such applications, image registration of the recorded sequence must be performed. Here we use a new non-mydriatic video-ophthalmoscope for simple and fast acquisition of low SNR retinal sequences. We introduce a novel, two-step approach for fast image registration. The phase correlation in the first stage removes large eye movements. Lucas-Kanade tracking in the second stage removes small eye movements. We propose robust adaptive selection of the tracking points, which is the most important part of trac...

2015 Computing in Cardiology Conference (CinC), 2015

Isolated cardiomyocytes have been used as valid and useful model in experimental cardiology resea... more Isolated cardiomyocytes have been used as valid and useful model in experimental cardiology research for decades. The cell contraction function is usually measured via expensive and complex instruments which can either damage the cell or take much time for setting up. In contrary, recent development of optical microscopy and digital cameras suggests utilization of touch-less cardiomyocyte video acquisition in connection with advanced image processing techniques for evaluation of the cell contraction process. The proposed paper presents an automatic membrane detection method via computer processing of acquired video-sequences by utilization of an active contour model. Evaluation of detected cell area is used for estimation of cardiomyocyte contraction function. The method is evaluated utilizing the comparison with contraction measurement performed via atomic force microscopy technique.

The European project EVICAB is devoted to implementing the idea of a virtual university campus fo... more The European project EVICAB is devoted to implementing the idea of a virtual university campus for teaching a standardized curriculum of Biomedical Engineering (BME). This joint effort of several universities from Northern and central European countries should lead to a possibility for students from anywhere in Europe (or even in the world) to study the Biomedical Engineering via Internet connection, still under guidance of hopefully some of the best BME tutors from different European universities offering quality elearning courses to EVICAB. It has turned out that one of the important issues to be solved in frame of the project is the design of a suitable organizational structure that would enable smooth running of the virtual university in all respects, i.e. tuition, home works and projects, and also reliable tests and exams. The main ideas of the corresponding organizational structure are described.