A5107305692- Medical Image Segmentation Techniques
- Advanced Image Processing Techniques
- Brain Tumor Detection and Classification
- Image and Signal Denoising Methods
- Advanced Data Compression Techniques
- Radiomics and Machine Learning in Medical Imaging
- Advanced Vision and Imaging
- Digital Media Forensic Detection
- Image Processing Techniques and Applications
- AI in cancer detection
- Face recognition and analysis
- Face and Expression Recognition
- Advanced Measurement and Detection Methods
- Advanced Scientific Research Methods
- MRI in cancer diagnosis
- Image Retrieval and Classification Techniques
- Advanced Neural Network Applications
- Infrared Target Detection Methodologies
- Advanced Steganography and Watermarking Techniques
- Advanced X-ray and CT Imaging
- Video Analysis and Summarization
- Cell Image Analysis Techniques
- Color Science and Applications
- Advanced MRI Techniques and Applications
- Renal and Vascular Pathologies
XLIM
2014-2025
Université de Poitiers
2016-2025
Imagerie par Résonance Magnétique Médicale et Multi-Modalités
2012-2025
Centre National de la Recherche Scientifique
2005-2024
Centre Hospitalier Universitaire de Poitiers
2021-2022
Siemens (France)
2020
Siemens (Germany)
2020
Laboratoire des Sciences de l’Information et de la Communication
2005-2014
Unité de Recherche Pluridisciplinaire Prairies et Plantes Fourragères
2014
Technicolor (France)
2008-2013
Accurate detection and quantification of unruptured intracranial aneurysms (UIAs) is important for rupture risk assessment to allow an informed treatment decision be made. Currently, 2D manual measures used assess UIAs on Time-of-Flight magnetic resonance angiographies (TOF-MRAs) lack 3D information there substantial inter-observer variability both aneurysm size growth. could helpful improve but are time-consuming would therefore benefit from a reliable automatic UIA segmentation method. The...
Images available on online sharing platforms have a high probability of being modified, with additional global transformations such as compression, resizing or filtering covering the possible alteration. Such manipulations impose many constraints forgery detection algorithms. This article presents framework improving robustness for image detection. The most important step our is to take into account quality corresponding chosen application. Therefore, we relied camera identification model...
Purpose: The automatic segmentation of multiple sclerosis lesions in magnetic resonance imaging has the potential to reduce radiologists' efforts on a daily time-consuming task and bring more reproducibility. Almost all new techniques make use convolutional neural networks with their own different architecture. Architectural choices are rarely explained. We aimed at presenting relevance U-net-like architecture for our specific building an efficient simple model. Approach: An experimental...
We propose a nonrigid registration method whose motion estimation is cast into feature matching problem under the Log-Demons framework using Graph Wavelets. investigate Spectral Wavelets (SGWs) to capture shape features of images. The SGWs are more adapted learn spatial and geometric organization data with complex structures than classical wavelets. Our experiments on T1 brain images endomicroscopic show that this outperforms existing image techniques (i.e. Log-Demons) improved similarity values.
The high-resolution magnetic resonance image (MRI) provides detailed anatomical information critical for clinical application diagnosis.However, MRI typically comes at the cost of long scan time, small spatial coverage, and low signal-to-noise ratio.The benefits convolutional neural network (CNN) can be applied to solve super-resolution task recover generic images from low-resolution inputs.Additionally, recent studies have shown potential use generative advertising (GAN) generate...
The prevention of cardiovascular diseases starts by a thorough examination the coronary artery vessels for atherosclerotic plaques existence. By combining deep learning convolutional Neural Network (CNN) architectures and biological knowledge, we introduce novel method automatic extraction centerlines in Computed Tomography Angiography (CTA) data. proposed is based on 3D neural network used as local vessel centerline detector to extract main side branches tree. Coupled with preprocessing...
This document proposes a convenient theoretical analysis of light modulation-based systems for prevention illegal recordings in movie theaters. Although the works presented this paper do not solve problem camcorder piracy, people security community may find them interesting further work area.
Glioma grade classification based on Magnetic Resonance (MR) data and using Machine Learning approaches is a hot topic. Recently, considerable improvements have been made in this field especially the last two years. This paper reviews selection of most recent methods from 2018 2019, details their preprocessing priors, such as different modalities used datasets. It then groups by comparing learning scheme. While classical machine present, more authors are Convolutional Neural Networks....
This document proposes a new, multi-primary projection system for prevention of illegal recordings in movie theaters. As today, most camcorder-jamming methods tend to rely on either Infra-Red or spatial/temporal light modulations, both being easy defeat using an appropriate filter system. Like Infra-Red, metamerism-based will modulate magnitude depending wavelength, but this time within the range visible light, making jamming patterns harder out. The mathematical model we use solve our...
Assessment of renal microstructure and function non-invasively, has an important role in monitoring predicting chronic kidney disease (CKD). The goal this study is to differentiate healthy CKD patients using texture analysis. Apparent diffusion coefficient (ADC) maps were generated from weighted magnetic resonance images (DWI), which statistical wavelet-transform based parameters extracted. results preliminary indicated that affects parameters. correlation the energy wavelet "low-high"...
In this paper, we propose a new polynomial based texture representation method for extracting information about facial expressions. While many appearance-based methods have been proposed over the years to improve performance of expression recognition, most descriptors are usually unable both provide precise multi-scale / multi-orientation analysis and handle redundancy problem effectively. We will explain how coefficients obtained from projections pixel intensities on complete basis can be...