FDG-PET to T1 Weighted MRI Translation with 3D Elicit Generative Adversarial Network (E-GAN).

OBJECTIVE: With the strengths of deep learning, computer-aided diagnosis (CAD) is a hot topic for researchers in medical image analysis. One of the main requirements for training a deep learning model is providing enough data for the network. However, in medical images, due to the difficulties of data collection and data privacy, finding an appropriate dataset (balanced, enough samples, etc.) is quite a challenge. Although image synthesis could be beneficial to overcome this issue, synthesizing 3D images is a hard task. The main objective of this paper is to generate 3D T1 weighted MRI corresponding to FDG-PET. In this study, we propose a separable convolution-based Elicit generative adversarial network (E-GAN). The proposed architecture can reconstruct 3D T1 weighted MRI from 2D high-level features and geometrical information retrieved from a Sobel filter. Experimental results on the ADNI datasets for healthy subjects show that the proposed model improves the quality of images compared with the state of the art. In addition, the evaluation of E-GAN and the state of art methods gives a better result on the structural information (13.73% improvement for PSNR and 22.95% for SSIM compared to Pix2Pix GAN) and textural information (6.9% improvements for homogeneity error in Haralick features compared to Pix2Pix GAN).

Surface Regularity via the Estimation of Fractional Brownian Motion Index.

The recent definition of fractional Brownian motions on surfaces has raised the statistical issue of estimating the Hurst index characterizing these models. To deal with this open issue, we propose a method which is based on a spectral representation of surfaces built upon their Laplace-Beltrami operator. This method includes a first step where the surface supporting the motion is recovered using a mean curvature flow, and a second one where the Hurst index is estimated by linear regression on the motion spectrum. The method is evaluated on synthetic surfaces. The interest of the method is further illustrated on some fetal cortical surfaces extracted from magnetic resonance images as a means to quantify the brain complexity during the gestational age.

Bayesian technique for image classifying registration.

In this paper, we address a complex image registration issue arising while the dependencies between intensities of images to be registered are not spatially homogeneous. Such a situation is frequently encountered in medical imaging when a pathology present in one of the images modifies locally intensity dependencies observed on normal tissues. Usual image registration models, which are based on a single global intensity similarity criterion, fail to register such images, as they are blind to local deviations of intensity dependencies. Such a limitation is also encountered in contrast-enhanced images where there exist multiple pixel classes having different properties of contrast agent absorption. In this paper, we propose a new model in which the similarity criterion is adapted locally to images by classification of image intensity dependencies. Defined in a Bayesian framework, the similarity criterion is a mixture of probability distributions describing dependencies on two classes. The model also includes a class map which locates pixels of the two classes and weighs the two mixture components. The registration problem is formulated both as an energy minimization problem and as a maximum a posteriori estimation problem. It is solved using a gradient descent algorithm. In the problem formulation and resolution, the image deformation and the class map are estimated simultaneously, leading to an original combination of registration and classification that we call image classifying registration. Whenever sufficient information about class location is available in applications, the registration can also be performed on its own by fixing a given class map. Finally, we illustrate the interest of our model on two real applications from medical imaging: template-based segmentation of contrast-enhanced images and lesion detection in mammograms. We also conduct an evaluation of our model on simulated medical data and show its ability to take into account spatial variations of intensity dependencies while keeping a good registration accuracy.

A local technique based on vectorized surfaces for craniofacial reconstruction.

In this paper, we focus on the automation of facial reconstruction. Since they consider the whole head as the object of interest, usual reconstruction techniques are global and involve a large number of parameters to be estimated. We present a local technique which aims at reaching a good trade-off between bias and variance following the paradigm of non-parametric statistics. The estimation is localized on patches delimited by surface geodesics between anatomical points of the skull. The technique relies on a continuous representation of the individual surfaces embedded in the vectorial space of extended normal vector fields. This allows to compute deformations and averages of surfaces. It consists in estimating the soft-tissue surface over patches. Using a homogeneous database described in [31], we obtain results on the chin and nasal regions with an average error below 1mm, outperforming the global reconstruction techniques.

A classifying registration technique for the estimation of enhancement curves of DCE-CT scan sequences.

In this paper, we propose a new technique for the estimation of contrast enhancement curves of Dynamic Contrast-Enhanced sequences, which takes the most from the interdependence between this estimation problem and the registration problem raised by possible movements occurring in sequences. The technique solves the estimation and registration problems simultaneously in an iterative way. However, unlike previous techniques, a pixel classification scheme is included within the estimation so as to compute enhancement curves on pixel classes instead of single pixels. The classification scheme is designed using a descendant hierarchical approach. Due to this tree approach, the number of classes is set automatically and the whole technique is entirely unsupervised. Moreover, some specific prior information about the shape of enhancement curves are included in the splitting and pruning steps of the classification scheme. Such an information ensures that created classes include pixels having homogeneous and relevant enhancement properties. The technique is applied to DET-CT scan sequences and evaluated using ground truth data. Results show that classifications are anatomically sound and that contrast enhancements are accurately estimated from sequences.

Mammogram registration: a phantom-based evaluation of compressed breast thickness variation effects.

The temporal comparison of mammograms is complex; a wide variety of factors can cause changes in image appearance. Mammogram registration is proposed as a method to reduce the effects of these changes and potentially to emphasize genuine alterations in breast tissue. Evaluation of such registration techniques is difficult since ground truth regarding breast deformations is not available in clinical mammograms. In this paper, we propose a systematic approach to evaluate sensitivity of registration methods to various types of changes in mammograms using synthetic breast images with known deformations. As a first step, images of the same simulated breasts with various amounts of simulated physical compression have been used to evaluate a previously described nonrigid mammogram registration technique. Registration performance is measured by calculating the average displacement error over a set of evaluation points identified in mammogram pairs. Applying appropriate thickness compensation and using a preferred order of the registered images, we obtained an average displacement error of 1.6 mm for mammograms with compression differences of 1-3 cm. The proposed methodology is applicable to analysis of other sources of mammogram differences and can be extended to the registration of multimodality breast data.