A5005539981- Advanced MRI Techniques and Applications
- Medical Imaging Techniques and Applications
- Medical Image Segmentation Techniques
- Radiomics and Machine Learning in Medical Imaging
- Cardiac Imaging and Diagnostics
- Advanced X-ray and CT Imaging
- Cardiovascular Health and Disease Prevention
- Retinal Imaging and Analysis
- Thermoregulation and physiological responses
- Spectroscopy and Chemometric Analyses
- Infrared Thermography in Medicine
- Cardiomyopathy and Myosin Studies
- Advanced Neural Network Applications
- Cardiovascular Function and Risk Factors
- Industrial Vision Systems and Defect Detection
- Complex Systems and Time Series Analysis
- Cardiovascular Disease and Adiposity
- Blind Source Separation Techniques
- Digital Imaging for Blood Diseases
- Electron and X-Ray Spectroscopy Techniques
- MRI in cancer diagnosis
- Cardiac Valve Diseases and Treatments
- Retinal and Optic Conditions
- Advanced Image and Video Retrieval Techniques
- Image Processing Techniques and Applications
Université de Bourgogne
2013-2024
Imagerie et Cerveau
2024
ImViA - Imagerie et Vision Artificielle
2021-2023
Université Bourgogne Franche-Comté
2018-2023
Laboratoire d’Électronique, Informatique et Image
2012-2019
Laboratoire Interuniversitaire des Systèmes Atmosphériques
2011-2018
Laboratoire Techniques, Territoires et Sociétés
2011-2018
Centre National de la Recherche Scientifique
2006-2018
Supélec
2013
Institut national de recherche en informatique et en automatique
2013
In recent years, several deep learning models have been proposed to accurately quantify and diagnose cardiac pathologies. These automated tools heavily rely on the accurate segmentation of structures in MRI images. However, right ventricle is challenging due its highly complex shape ill-defined borders. Hence, there a need for new methods handle such structure's geometrical textural complexities, notably presence pathologies as Dilated Right Ventricle, Tricuspid Regurgitation,...
Deep learning-based methods for cardiac MR segmentation have achieved state-of-the-art results. However, these can generate incorrect results which lead to wrong clinical decisions in the downstream tasks. Automatic and accurate analysis of tasks, such as myocardial tissue characterization, is highly dependent on quality Therefore, it paramount importance use control detect failed segmentations before further analysis. In this work, we propose a fully automatic uncertainty-based framework T1...
Accurate segmentation of the myocardial scar may supply relevant advancements in predicting and controlling deadly ventricular arrhythmias subjects with cardiovascular disease. In this paper, we propose architecture inclusion classification prior information U-Net (ICPIU-Net) to efficiently segment left ventricle (LV) myocardium, infarction (MI), microvascular-obstructed (MVO) tissues from late gadolinium enhancement magnetic resonance (LGE-MR) images. Our approach was developed using two...
To propose, assess, and validate a semiautomatic method allowing rapid reproducible measurement of trabeculated compacted left ventricular (LV) masses from cardiac magnetic resonance imaging (MRI).We developed to automatically detect noncompacted, endocardial, epicardial contours. Papillary muscles were segmented using thresholding included in the mass. Blood was removed trabeculae same threshold tool. Trabeculated, ratio noncompacted (NC:C) computed. Preclinical validation performed on four...
In this paper, we present a new Markovian scheme for MRI segmentation using priori knowledge obtained from probability maps. Indeed propose to use both triplet Markov chain and brain atlas containing prior expectations about the spatial localization of different tissue classes, segment in gray matter, white matter cerebro-spinal fluid an unsupervised way. Experimental results on real data are included validate approach. Comparison with other previously used techniques demonstrates advantages...
this paper, we present a new automatic robust algorithm to segment multimodal brain MR images with Multiple Sclerosis (MS) lesions. The method performs tissue classification using Hidden Markov Chain (HMC) model and detects MS lesions as outliers the model. For aim, use Trimmed Likelihood Estimator (TLE) extract outliers. Furthermore, neighborhood information is included HMC propose incorporate priori brought by probabilistic atlas. Tests on Brainweb have been carried out validate approach.
This paper proposes a new method to detect multiple sclerosis (MS) lesions on 3D multimodal brain MR images. MS are detected as voxels that not well explained by statistical model for normal These outliers extracted using the trimmed likelihood estimator (TLE). Spatial regularization is performed hidden Markov chain (HMC) model. Tests real images with have been carried out and results compared manual expert segmentation validate proposed method.
Tissue segmentation and classification in MRI is a challenging task due to lack of signal intensity standardization. dependent on the acquisition protocol, coil profile, scanner type, etc. While we can compute quantitative physical tissue properties independent hardware sequence parameters, it still difficult leverage these segment classify pelvic tissues. The proposed method integrates values (T1 T2 relaxation times pure synthetic weighted images) machine learning (Support Vector Machine...
Precision characterization is fundamental to achieve expected performance in semiconductors where Moore's law pushes the boundaries miniaturize components. To measure these attributes, deep learning models are used, which require manual annotation of several objects captured via electron microscopy. However, this can be laborious and time-consuming. We propose a semi-automated method for annotating items microscopy images, an effort innovative, efficient, reliable. Our approach involves...
Deep learning models have been proven to automate metrology tasks. It provides accurate, robust and fast results if it is trained with proper data. Nonetheless, obtaining training data remains tedious. requires an expert user delimitate objects boundaries in several images representing tens hundreds of objects. Instead drawing precise boundaries, we propose a tool relying on rectangular bounding box detect segment For complex applications non-homogeneous background, the must draw one per...
In this paper, we present a new automatic robust algorithm to segment multimodal brain MR images with Multiple Sclerosis (MS) lesions. The method performs tissue classification using Hidden Markov Chain (HMC) model and detects MS lesions as outliers the model. For aim, use Trimmed Likelihood Estimator (TLE) extract outliers. Furthermore, neighborhood information is included HMC propose incorporate priori brought by probabilistic atlas.
In this paper, we present a robust method to estimate parameters of hidden Markov chains (HMC) in order segment brain MR images. Indeed, parameter estimation can be very sensitive the presence outliers data. We propose use trimmed likelihood estimator (TLE) extract such and accurately different tissue classes way. Moreover neighborhood information is included model by using chains. Experimental results on 2D synthetic data 3D MRI are validate approach.