arXiv subjects

Alain Lalande

Publications and source records attributed to Alain Lalande.

22 records · Page 2Linked to original sources

arXiv2018

Sigmoid function based intensity transformation for parameter initialization in MRI-PET Registration Tool for Preclinical Studies

Images from Positron Emission Tomography (PET) deliver functional data such as perfusion and metabolism. On the other hand, images from Magnetic Resonance Imaging (MRI) provide information describing anatomical structures. Fusing the complementary information from the two modalities is helpful in oncology. In this project, we implemented a complete tool allowing semi-automatic MRI-PET registration for small animal imaging in the preclinical studies. A two-stage hierarchical registration approach is proposed. First, a global affine registration is applied. For robust and fast registration, principal component analysis (PCA) is used to compute the initial parameters for the global affine registration. Since, only the low intensities in the PET volume reveal the anatomic information on the MRI scan, we proposed a non-uniform intensity transformation to the PET volume to enhance the contrast of the low intensity. This helps to improve the computation of the centroid and principal axis by increasing the contribution of the low intensities. Then, the globally registered image is given as input to the second stage which is a local deformable registration (B-spline registration). Mutual information is used as metric function for the optimization. A multi-resolution approach is used in both stages. The registration algorithm is supported by graphical user interface (GUI) and visualization methods so that the user can interact easily with the process. The performance of the registration algorithm is validated by two medical experts on seven different datasets on abdominal and brain areas including noisy and difficult image volumes.

eess.IV↗

arXiv2017

GridNet with automatic shape prior registration for automatic MRI cardiac segmentation

In this paper, we propose a fully automatic MRI cardiac segmentation method based on a novel deep convolutional neural network (CNN) designed for the 2017 ACDC MICCAI challenge. The novelty of our network comes with its embedded shape prior and its loss function tailored to the cardiac anatomy. Our model includes a cardiac centerof-mass regression module which allows for an automatic shape prior registration. Also, since our method processes raw MR images without any manual preprocessing and/or image cropping, our CNN learns both high-level features (useful to distinguish the heart from other organs with a similar shape) and low-level features (useful to get accurate segmentation results). Those features are learned with a multi-resolution conv-deconv "grid" architecture which can be seen as an extension of the U-Net. Experimental results reveal that our method can segment the left and right ventricles as well as the myocardium from a 3D MRI cardiac volume in 0.4 second with an average Dice coefficient of 0.90 and an average Hausdorff distance of 10.4 mm.

cs.CV↗

arXiv2017

Semi-automated labelling of medical images: benefits of a collaborative work in the evaluation of prostate cancer in MRI

Purpose: The goal of this study is to show the advantage of a collaborative work in the annotation and evaluation of prostate cancer tissues from T2-weighted MRI compared to the commonly used double blind evaluation. Methods: The variability of medical findings focused on the prostate gland (central gland, peripheral and tumoural zones) by two independent experts was firstly evaluated, and secondly compared with a consensus of these two experts. Using a prostate MRI database, experts drew regions of interest (ROIs) corresponding to healthy prostate (peripheral and central zones) and cancer using a semi-automated tool. One of the experts then drew the ROI with knowledge of the other expert's ROI. Results: The surface area of each ROI as the Hausdorff distance and the Dice coefficient for each contour were evaluated between the different experiments, taking the drawing of the second expert as the reference. The results showed that the significant differences between the two experts became non-significant with a collaborative work. Conclusions: This study shows that collaborative work with a dedicated tool allows a better consensus between expertise than using a double blind evaluation. Although we show this for prostate cancer evaluation in T2-weighted MRI, the results of this research can be extrapolated to other diseases and kind of medical images.

physics.med-ph↗

arXiv2017

MRI-PET Registration with Automated Algorithm in Pre-clinical Studies

Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) automatic 3-D registration is implemented and validated for small animal image volumes so that the high-resolution anatomical MRI information can be fused with the low spatial resolution of functional PET information for the localization of lesion that is currently in high demand in the study of tumor of cancer (oncology) and its corresponding preparation of pharmaceutical drugs. Though many registration algorithms are developed and applied on human brain volumes, these methods may not be as efficient on small animal datasets due to lack of intensity information and often the high anisotropy in voxel dimensions. Therefore, a fully automatic registration algorithm which can register not only assumably rigid small animal volumes such as brain but also deformable organs such as kidney, cardiac and chest is developed using a combination of global affine and local B-spline transformation models in which mutual information is used as a similarity criterion. The global affine registration uses a multi-resolution pyramid on image volumes of 3 levels whereas in local B-spline registration, a multi-resolution scheme is applied on the B-spline grid of 2 levels on the finest resolution of the image volumes in which only the transform itself is affected rather than the image volumes. Since mutual information lacks sufficient spatial information, PCA is used to inject it by estimating initial translation and rotation parameters. It is computationally efficient since it is implemented using C++ and ITK library, and is qualitatively and quantitatively shown that this PCA-initialized global registration followed by local registration is in close agreement with expert manual registration and outperforms the one without PCA initialization tested on small animal brain and kidney.

cs.CV↗