Evaluation of cortical segmentation pipelines on clinical neonatal MRI data.

Magnetic Resonance Imaging (MRI) can provide 3D morphological information on brain structures. Such information is particularly relevant for carrying out morphometric brain analysis, especially in the newborn and in the case of prematurity. However, 3D neonatal MRI acquired in clinical environments are low-resolution, anisotropic images, making segmentation a challenging task. In this context, preprocessing techniques aim to increase the image resolution. Interpolation techniques were classically used; super-resolution (SR) techniques have recently appeared as an emerging alternative. In this paper, we evaluate the performance of different SR methods against the classical interpolation in the application of neonatal cortex segmentation. Additionally, we assess the robustness of different segmentation methods for each estimation of high resolution MRI input. Results are evaluated both qualitatively and quantitatively with neonatal clinical MRI.

Segmentation of Mueller matrix images under non-uniform illumination.

The recently proposed restoration-segmentation algorithms dedicated to polarization encoded images suffer two important limitations: the number of classes into which the image is segmented is not obtained automatically, and more importantly the quality of the segmentation is affected by the nonuniformity of the illumination of the scene. We propose here a new method addressing these issues. It is based on a global estimation-segmentation approach, explicitly modeling the nonuniform illumination. The physical admissibility of the retrieved Mueller matrices is ensured. Results stemming from synthetic and real data are provided and support the proposed approach.

Scanning ophthalmoscope retinal image registration using one-dimensional deformation fields.

We present a new, robust and automated method for registering sequences of images acquired from scanning ophthalmoscopes. The method uses a multi-scale B-spline representation of the deformation field to map images to each other and an hierarchical optimization method. We applied the method to video sequences acquired from different parts of the retina. In all cases, the registration was successful, even in the presence of large distortions from microsaccades, and the resulting deformation fields describe the fixational motion of the eye. The registration reveals examples of dynamic photoreceptor behaviour in the sequences.

An a contrario approach for change detection in 3D multimodal images: application to multiple sclerosis in MRI.

Estimating significant changes between two images remains a challenging problem in medical image processing. This paper proposes a non-parametric region based method to detect significant changes in 3D multimodal Magnetic Resonance (MR) sequences. The proposed approach relies on an a contrario model which defines significant changes as events with very low probability. We adapt the a contrario framework to deal with multimodal images from which are extracted measures related to intensity and volume changes. Two fusion rules are carefully designed to handle a set of decision thresholds and a set of image measures. The final decision is taken using multiple testing procedures. The efficiency of the algorithm is demonstrated in the context of multiple sclerosis (MS) lesion analysis over time in multimodal MR sequences. We evaluate the proposed method on synthetic images using the Brainweb simulator. Finally, promising results on multimodal sequences on clinical data are presented.

Hidden Markov multiple event sequence models: A paradigm for the spatio-temporal analysis of fMRI data.

This paper presents a novel, completely unsupervised fMRI brain mapping method that addresses the three problems of hemodynamic response function (HRF) variability, hemodynamic event timing, and fMRI response non-linearity. Spatial and temporal information are directly taken into account into the core of the activation detection process. In practice, activation detection at voxel v is formulated in terms of temporal alignment between sequences of hemodynamic response onsets (HROs) detected in the fMRI signal at v and in the spatial neighborhood of v, and the input sequence of stimuli or stimulus onsets. Event-related and epoch paradigms are considered. The multiple event sequence alignment problem is solved within the probabilistic framework of hidden Markov multiple event sequence models (HMMESMs), a new class of hidden Markov models. Results obtained on real and synthetic data significantly outperform those obtained with the popular statistical parametric mapping (SPM2) method without requiring any prior definition of the expected activation patterns, the HMMESM mapping approach being completely unsupervised.