A5067448884- Medical Imaging Techniques and Applications
- Advanced X-ray and CT Imaging
- Radiomics and Machine Learning in Medical Imaging
- Medical Image Segmentation Techniques
- Digital Image Processing Techniques
- Advanced Radiotherapy Techniques
- Radiation Dose and Imaging
- AI in cancer detection
- MRI in cancer diagnosis
- Cultural, Linguistic, Economic Studies
- Image and Signal Denoising Methods
- Topological and Geometric Data Analysis
- Advanced Optical Sensing Technologies
- Computer Graphics and Visualization Techniques
- Radiopharmaceutical Chemistry and Applications
- Advanced Vision and Imaging
- Advanced Image Processing Techniques
- Advanced Numerical Analysis Techniques
- Acute Ischemic Stroke Management
- Sparse and Compressive Sensing Techniques
- Cerebrovascular and Carotid Artery Diseases
- Advanced MRI Techniques and Applications
- Image and Video Quality Assessment
Centrum Wiskunde & Informatica
2018-2021
College of Western Idaho
2019
Centre National de la Recherche Scientifique
2013-2016
Institut Génétique Nantes Atlantique
2012-2016
Bellingham Technical College
2016
Nantes Université
2016
Unlike previous works, this open data collection consists of X-ray cone-beam (CB) computed tomography (CT) datasets specifically designed for machine learning applications and high cone-angle artefact reduction. Forty-two walnuts were scanned with a laboratory set-up to provide not only from single object but class objects natural variability. For each walnut, CB projections on three different source orbits acquired cone angles as well being able compute artefact-free, high-quality ground...
Current computational methods for light field photography model the ray-tracing geometry inside plenoptic camera. This representation of problem, and some common approximations, can lead to errors in estimation object sizes positions. We propose a that leads correct reconstruction distances camera, by showing images be interpreted as limited angle cone-beam tomography acquisitions. then quantitatively analyze its impact on image refocusing, depth volumetric reconstructions, comparing it...
Two new methods to perform interpolation mapping from Radon sinogram Mojette domain are presented. Reconstructions made both spaces using FBP and SART algorithms. Assessment of the is Shepp-Logan phantom actual data demonstrate efficiency proposed
The ability to predict tumor recurrence after chemoradiotherapy of locally advanced cervical cancer is a crucial clinical issue intensify the treatment most high-risk patients. objective this study was investigate metabolism characteristics extracted from pre- and per-treatment <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">18</sup> F-FDG PET images 3-year overall (OR). A total 53 patients underwent (respectively PET1 PET2). Tumor characterized...
In this paper, we develop an exact, reversible and scale change rotation in the Mojette projection space. The whole process is performed using 1D fast operators has advantage to be consistent with standard tomographic geometry.
Patient-specific dosimetry in nuclear medicine relies on activity quantification volumes of interest from scintigraphic imaging. Clinical protocols have to be benchmarked against results computed test phantoms. The design an adequate model is a crucial step for the validation image-based activ ity quantification. We propose computing platform automatically generate simulated SPECT images dynamic phantom arbitrary image protocols. As regards generation, we first use open-source NCAT code...
Computed Tomography (CT) is an imaging technique that allows to reconstruct volumetric information of the analyzed objects from their projections. The most popular reconstruction Filtered Back Projection (FBP). It has advantage being fastest available, but also disadvantage require a high number projections retrieve good quality reconstructions. In this article we propose segmentation method for tomographic volumes composed few materials. Our combines existing high-quality variational...
Micro-CT represents a modality where the quality of CT reconstruction is very high thanks to acquisition properties. The goal this paper challenge our proposed Mojette discrete scheme from real micro-CT data. A first study was done analyze bone image degradations by lowering number projections. second analyzes trabecular and vessels tree through an animal study. Small are filling holes with almost same grey levels as bone. Therefore vessel detectability that can be achieved algorithm...
The Mojette transform is a discrete and exact Radon transform, based on the geometry of projection reconstruction lattice. specific sampling scheme results in theoretical image reconstruction. In this paper, we compare reconstructions obtained with to ones several usual projection/backprojection digitized transform. These experiments validate demonstrate performance over classical implementations continuous space.
This paper describes the framework for creation of whole-body planar acquisitions from Monte Carlo modelling with GATE. The ground truth is a complex model representing `virtual patient' based on NCAT-WB anthropomorphic model. Radiopharmaceutical kinetic was generated compartmental modelling, to assign time-activity curve (TAC) each functional compartment. In order match model's compartments, geometrical used define corresponding volumes, where geometric voxel results linear combination...
High cone-angle artifacts (HCAAs) appear frequently in circular cone-beam computed tomography (CBCT) images and can heavily affect diagnosis treatment planning. To reduce HCAAs CBCT scans, we propose a novel deep learning approach that reduces the three-dimensional (3D) nature of to two-dimensional (2D) problems an efficient way. Specifically, exploit relationship between rotational scanning geometry by training convolutional neural network (CNN) using image slices were radially sampled from...
In this paper, we develop a global iterative algorithm for tomographic reconstructions from Mojette projections. Since Spline-Mojette projections are obtained by convolving Dirac-Mojette values with specific uniform projection kernel, decorrelate model and provide scheme available all models. We refer algorithms to their Radon based counterparts propose comparative study several acquisitions.
Carotid surgery is a frequent act corresponding to 15 20 thousands operations per year in France. Cerebral perfusion has be tracked before and after carotid surgery. In this paper, diagnosis support using quality metrics proposed detect vascular lesions on MR images. Our key stake provide detection tool mimicking the human visual system behavior during inspection. Relevant Human Visual System (HVS) properties should integrated our lesion method, which must robust common distortions medical...