A5012732556- Medical Image Segmentation Techniques
- Advanced Image and Video Retrieval Techniques
- Image Retrieval and Classification Techniques
- Medical Imaging Techniques and Applications
- Cell Image Analysis Techniques
- AI in cancer detection
- Ultrasound Imaging and Elastography
- Image and Signal Denoising Methods
- Robotics and Sensor-Based Localization
- Electromagnetic Scattering and Analysis
- Image and Object Detection Techniques
- Advanced Vision and Imaging
- Radiomics and Machine Learning in Medical Imaging
- Electrical and Bioimpedance Tomography
- Sparse and Compressive Sensing Techniques
- Pancreatic function and diabetes
- Cerebrovascular and Carotid Artery Diseases
- Image Processing Techniques and Applications
- Digital Imaging for Blood Diseases
- Elasticity and Material Modeling
- Advanced MRI Techniques and Applications
- Acoustic Wave Phenomena Research
- Algorithms and Data Compression
- Advanced Image Processing Techniques
- Coronary Interventions and Diagnostics
Czech Technical University in Prague
2015-2024
Universidad Carlos III de Madrid
2018
PATH To Reading
2009-2012
Inserm
2010
École Polytechnique Fédérale de Lausanne
1999-2010
Centre de Recherche en Acquisition et Traitement de l'Image pour la Santé
2008-2010
Université Claude Bernard Lyon 1
2008
Institut national de recherche en informatique et en automatique
2002-2005
We present an algorithm for fast elastic multidimensional intensity-based image registration with a parametric model of the deformation. It is fully automatic in its default mode operation. In case hard real-world problems, it capable accepting expert hints form soft landmark constraints. Much fewer landmarks are needed and results far superior compared to pure registration. Particular attention has been paid factors influencing speed this algorithm. The B-spline deformation shown be...
The forward electroencephalography (EEG) problem involves finding a potential V from the Poisson equation /spl nabla//spl middot/(/spl sigma//spl nabla/V)=f, in which f represents electrical sources brain, and sigma/ conductivity of head tissues. In piecewise constant model, this can be accomplished by boundary element method (BEM) using suitable integral formulation. Most previous work uses same formulation, corresponding to double-layer potential. We present conceptual framework based on...
We propose a new spatio-temporal elastic registration algorithm for motion reconstruction from series of images. The specific application is to estimate displacement fields two-dimensional ultrasound sequences the heart. basic idea find deformation field that effectively compensates by minimizing difference with respect reference frame. key feature our method use semi-local parametric model using splines, and reformulation task as global optimization problem. scale spline controls smoothness...
Purpose: Four‐dimensional computed tomography (4D CT) can provide patient‐specific motion information for radiotherapy planning and delivery. Motion estimation in 4D CT is challenging due to the reduced image quality presence of artifacts. We aim improve robustness deformable registration applied respiratory‐correlated imaging lungs, by using a global problem formulation pursuing restrictive parametrization spatiotemporal deformation model. Methods: A spatial transformation based on...
Automatic Non-rigid Histological Image Registration (ANHIR) challenge was organized to compare the performance of image registration algorithms on several kinds microscopy histology images in a fair and independent manner. We have assembled 8 datasets, containing 355 with 18 different stains, resulting 481 pairs be registered. accuracy evaluated using manually placed landmarks. In total, 256 teams registered for challenge, 10 submitted results, 6 participated workshop. Here, we present...
Echo-planar imaging (EPI) is a fast nuclear magnetic resonance (MRI) method. Unfortunately, local field inhomogeneities induced mainly by the subject's presence cause significant geometrical distortion, predominantly along phase-encoding direction, which must be undone to allow for meaningful further processing. So far, this aspect has been too often neglected. In paper, we suggest new approach using an algorithm specifically developed automatic registration of distorted EPI images with...
We address the problem of estimating uncertainty pixel based image registration algorithms, given just two images to be registered, for cases when no ground truth data is available. Our novel method uses bootstrap resampling. It very general, applicable almost any on minimizing a pixel-based similarity criterion; we demonstrate it using SSD, SAD, correlation, and mutual information criteria. show experimentally that provides better estimates accuracy than state-of-the-art Cramer-Rao bound...
Ultrasound guidance is used for many surgical interventions such as biopsy and electrode insertion. We present a method to localize thin tool needle or microelectrode in 3-D ultrasound image. The proposed starts with thresholding model fitting using random sample consensus robust localization of the axis. Subsequent local optimization refines its position. Two different image models are presented: one simple fast second uses learned priori information about tool's voxel intensities...
Purpose: Deformable registration generally relies on the assumption that sought spatial transformation is smooth. Yet, breathing motion involves sliding of lung with respect to chest wall, causing a discontinuity in field, and smoothness can lead poor matching accuracy. In response, alternative methods have been proposed, several which rely prior segmentations. We propose an original method for automatically extracting particular segmentation, called mask, from CT image thorax. Methods: The...
Abstract Objective The objective of this study was to compare the detection caries in bitewing radiographs by multiple dentists with an automatic method and evaluate performance absence a reliable ground truth. Materials methods Four experts three novices marked using bounding boxes 100 radiographs. same dataset processed object deep learning method. All annotators were compared terms number errors intersection over union (IoU) pairwise comparisons, respect consensus standard, annotator...
Accurate geometrical models of the head are necessary for solving forward and inverse problems magneto- electro-encephalography (MEG/EEG). Boundary element methods (BEMs) require a model describing interfaces between different tissue types. Classically, with nested volume topology have been used. In this paper, we demonstrate how constraint can be relaxed, allowing us to more realistic topologies. We describe symmetric BEM new model. The formulation uses both potentials currents at as...
<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> In this paper, a 2-D locally regularized strain estimation method for imaging deformation of soft biological tissues from radio-frequency (RF) ultrasound (US) data is introduced. Contrary to most techniques that model the compression-induced local displacement as shift, our algorithm also considers scaling factor in axial direction. This direction-dependent tissue motion and induced by highly...
We present a new approach for matching sets of branching curvilinear structures that form graphs embedded in <inline-formula><tex-math notation="LaTeX">${\mathbb {R}}^2$</tex-math></inline-formula> or <inline-formula> <tex-math {R}}^3$</tex-math></inline-formula> and may be subject to deformations. Unlike earlier methods, ours does not rely on local appearance similarity nor require good initial alignment. Furthermore, it can cope with non-linear deformations, topological differences,...
Image registration is a common task for many biomedical analysis applications. The present work focuses on the benchmarking of methods differently stained histological slides. This challenging due to differences in appearance model, repetitive texture details and large image size, between other issues. Our data composed 616 pairs at two different scales - average diagonal 2.4k 5k pixels. We compare eleven fully automatic covering widely used similarity measures (and optimization strategies...
In surgical practice, small metallic instruments are frequently used to perform various tasks inside the human body. We address problem of their accurate localization in tissue. Recent experiments using medical ultrasound have shown that this modality is suitable for real-time visualization anatomical structures as well position instruments. propose an image- processing algorithm permits automatic estimation a line-segment-shaped object. This method was applied thin electrode biological show...
The accurate solution of the forward electrostatic problem is an essential first step before solving inverse magneto- and electroencephalography (MEG/EEG). symmetric Galerkin boundary element method but cannot be used for very large problems because its computational complexity memory requirements. We describe a fast multipole-based acceleration (BEM). It creates hierarchical structure elements approximates far interactions using spherical harmonics expansions. accelerated shown to as direct...
We present an efficient matching method for generalized geometric graphs. Such graphs consist of vertices in space connected by curves and can represent many real world structures such as road networks remote sensing, or vessel medical imaging. Graph be used very fast possibly multimodal registration images these structures. formulate the problem a single player game solved using Monte Carlo Tree Search, which automatically balances exploring new possible matches extending existing matches....
The number of international benchmarking competitions is steadily increasing in various fields machine learning (ML) research and practice. So far, however, little known about the common practice as well bottlenecks faced by community tackling questions posed. To shed light on status quo algorithm development specific field biomedical imaging analysis, we designed an survey that was issued to all participants challenges conducted conjunction with IEEE ISBI 2021 MICCAI conferences (80 total)....
We consider the problem of reconstructing a multidimensional vector function f/sub in/: /spl Ropf//sup m//spl rarr//spl n/ from finite set linear measures. These can be irregularly sampled responses several filters. Traditional approaches reconstruct in an priori given space, e.g., space bandlimited functions. Instead, we have chosen to specify reconstruction that is optimal sense quadratic plausibility criterion J. First, present solution generalized interpolation problem. Later, also...