Search

[EN] The human cerebellum plays an essential role in motor control, is involved in cognitive function (i.e., attention, working memory, and language), and helps to regulate emotional responses. Quantitative in-vivo assessment of the cerebellum is important in the study of several neurological diseases including cerebellar ataxia, autism, and schizophrenia. Different structural subdivisions of the cerebellum have been shown to correlate with differing pathologies. To further understand these pathologies, it is helpful to automatically parcellate the cerebellum at the highest fidelity possible. In this paper, we coordinated with colleagues around the world to evaluate automated cerebellum parcellation algorithms on two clinical cohorts showing that the cerebellum can be parcellated to a high accuracy by newer methods. We characterize these various methods at four hierarchical levels: coarse (i.e., whole cerebellum and gross structures), lobe, subdivisions of the vermis, and the lobules. Due to the number of labels, the hierarchy of labels, the number of algorithms, and the two cohorts, we have restricted our analyses to the Dice measure of overlap. Under these conditions, machine learning based methods provide a collection of strategies that are efficient and deliver parcellations of a high standard across both cohorts, surpassing previous work in the area. In conjunction with the rank-sum computation, we identified an overall winning method.

"This is the peer reviewed version of the following article: Planche, V., Koubiyr, I., Romero, J. E., Manjon, J. V., Coupé, P., Deloire, M., ... & Tourdias, T. (2018). Regional hippocampal vulnerability in early multiple sclerosis: Dynamic pathological spreading from dentate gyrus to CA 1. Human brain mapping, 39(4), 1814-1824., which has been published in final form at https://doi.org/10.1002/hbm.23970. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-Archiving.", [EN] Background: Whether hippocampal subfields are differentially vulnerable at the earliest stages of multiple sclerosis (MS) and how this impacts memory performance is a current topic of debate. Method: We prospectively included 56 persons with clinically isolated syndrome (CIS) suggestive of MS in a 1-year longitudinal study, together with 55 matched healthy controls at baseline. Participants were tested for memory performance and scanned with 3T MRI to assess the volume of 5 distinct hippocampal subfields using automatic segmentation techniques. Results: At baseline, CA4/dentate gyrus was the only hippocampal subfield with a volume significantly smaller than controls (p < .01). After one year, CA4/dentate gyrus atrophy worsened (-6.4%, p < .0001) and significant CA1 atrophy appeared (both in the stratum-pyramidale and the stratum radiatum-lacunosum-moleculare, -5.6%, p < .001 and -6.2%, p < .01, respectively). CA4/dentate gyrus volume at baseline predicted CA1 volume one year after CIS (R-2 = 0.44 to 0.47, p < .001, with age, T2 lesion-load, and global brain atrophy as covariates). The volume of CA4/dentate gyrus at baseline was associated with MS diagnosis during follow-up, independently of T2-lesion load and demographic variables (p < .05). Whereas CA4/dentate gyrus volume was not correlated with memory scores at baseline, CA1 atrophy was an independent correlate of episodic verbal memory performance one year after CIS (beta = 0.87, p < .05). Conclusion: The hippocampal degenerative process spread from dentate gyrus to CA1 at the earliest stage of MS. This dynamic vulnerability is associated with MS diagnosis after CIS and will ultimately impact hippocampal-dependent memory performance.

En la actualidad, la medicina y en especial el área de imagen médica, ha sido uno de los campos de la ciencia que más se ha beneficiado de las tecnologías de la información. En esta tesis nos centramos en las imágenes de resonancia magnética cerebral y los métodos de segmentación automática del volumen cerebral. En el primer capítulo describimos los antecedentes de este trabajo mediante una breve introducción de los principios físicos de la resonancia magnética y una revisión del estado del arte en relación con las técnicas de segmentación y su evolución a lo largo de las últimas décadas. En los capítulos tres, cuatro y cinco presentamos tres métodos de segmentación automática de diferentes partes del cerebro que mejoran el estado del arte en términos de calidad del resultado y velocidad. Finalmente, en el capítulo seis se comentan las conclusiones generales y se proponen líneas de desarrollo futuras., Nowadays, medicine and specially the medical image area, is one of the science fields that has benefited more from the information technologies. In this thesis we focus on cerebral magnetic resonance imaging and automatic segmentation methods of the brain volume. In the first chapter we describe a background for this work by a brief introduction of the physic fundaments of magnetic resonance and a revision of the state-of-the-art regarding to segmentation techniques and its evolution along the last decades. In the chapters three, four and five we present three methods for automatically segment different parts of the brain that leverage the state-of-the-art in terms of result quality and computation time. Finally, in chapter six we comment the general conclusions and propose lines for future works., En l'actualitat, la medicina i en especial l'àrea d'imatge mèdica, ha estat un dels camps de la ciència que més s'ha beneficiat de les tecnologies de la informació. En aquesta tesi ens centrem en les imatges de ressonància magnètica cerebral i el desenvolupament de mètodes de segmentaciò automàtica del volum cerebral. En el primer capítol descrvim els antecedents d'aquest treball mitjançant una breu introducció dels principis físics de la ressonància magnètica i una revisió de l'estat de l'art en relació amb les tècniques de segmentació i la seva evolució al llarg de les últimes dècades. En els capítols tres, quatre, i cinc presentem tres mètodes de segmentació automàtica per a diferents parts del cervell que milloren l'estat de l'art en quant a qualitat dels seus resultats i velocitat. Finalment, en el capitol sis es comenten les conclusions generals i es proposen línies de desenvolupament ions generals i es proposen línies de desenvolupament futures.

[EN] The importance of the hippocampus in the study of several neurodegenerative diseases such as Alzheimer's disease makes it a structure of great interest in neuroimaging. However, few segmentation methods have been proposed to measure its subfields due to its complex structure and the lack of high resolution magnetic resonance (MR) data. In this work, we present a new pipeline for automatic hippocampus subfield segmentation using two available hippocampus subfield delineation protocols that can work with both high and standard resolution data. The proposed method is based on multi-atlas label fusion technology that benefits from a novel multi-contrast patch match search process (using high resolution T1-weighted and T2-weighted images). The proposed method also includes as post-processing a new neural network-based error correction step to minimize systematic segmentation errors. The method has been evaluated on both high and standard resolution images and compared to other state-of-the-art methods showing better results in terms of accuracy and execution time.

[EN] The human cerebellum is involved in language, motor tasks and cognitive processes such as attention or emotional processing. Therefore, an automatic and accurate segmentation method is highly desirable to measure and understand the cerebellum role in normal and pathological brain development. In this work, we propose a patch-based multi-atlas segmentation tool called CERES (CEREbellum Segmentation) that is able to automatically parcellate the cerebellum lobules. The proposed method works with standard resolution magnetic resonance T1-weighted images and uses the Optimized PatchMatch algorithm to speed up the patch matching process. The proposed method was compared with related recent state-of-the-art methods showing competitive results in both accuracy (average DICE of 0.7729) and execution time (around 5 minutes).

"NOTICE: this is the author’s version of a work that was accepted for publication in Magnetic Resonance Imaging. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Magnetic Resonance Imaging, [Volume 33, Issue 4, May 2015, Pages 474–484] DOI 10.1016/j.mri.2015.02.005, In this paper, we propose an automatic method to segment the five main brain sub-regions (i.e. left/right hemispheres, left/right cerebellum and brainstem) from magnetic resonance images. The proposed method uses a library of pre-labeled brain images in a stereotactic space in combination with a non-local label fusion scheme for segmentation. The main novelty of the proposed method is the use of a multi-label block-wise label fusion strategy specifically designed to deal with the classification of main brain sub-volumes that process only specific parts of the brain images significantly reducing the computational burden. The proposed method has been quantitatively evaluated against manual segmentations. The evaluation showed that the proposed method was faster while producing more accurate segmentations than a current state-of-the-art method. We also present evidences suggesting that the proposed method was more robust against brain pathologies than the compared method. Finally, we demonstrate the clinical value of our method compared to the state-of-the-art approach in terms of the asymmetry quantification in Alzheimer's disease.

[EN] Automatic and accurate methods to estimate normalized regional brain volumes from MRI data are valuable tools which may help to obtain an objective diagnosis and followup of many neurological diseases. To estimate such regional brain volumes, the intracranial cavity volume (ICV) is often used for normalization. However, the high variability of brain shape and size due to normal intersubject variability, normal changes occurring over the lifespan, and abnormal changes due to disease makes the ICV estimation problem challenging. In this paper, we present a new approach to perform ICV extraction based on the use of a library of prelabeled brain images to capture the large variability of brain shapes. To this end, an improved nonlocal label fusion scheme based on BEaST technique is proposed to increase the accuracy of the ICV estimation. The proposed method is compared with recent state-of-the-art methods and the results demonstrate an improved performance both in terms of accuracy and reproducibility while maintaining a reduced computational burden.

El estudio del volumen intracraneal requiere del uso de herramientas que permitan objetivar el diagnóstico y ofrezcan un rendimiento y precisión elevados. La segmentación automática del volumen cerebral es el primer paso hacia un estudio más completo del cerebro y supondrá una herramienta versátil en el estudio de diversas patologías. Propósito: Mejorar un método ya implementado que segmenta el volumen cerebral con una calidad aceptable pero en un tiempo de ejecución elevado basado en comparación de regiones contra una biblioteca de casos de ejemplo segmentada manualmente. Método: El método recorre uno a uno todos los voxels del cerebro a segmentar extrayendo la región que lo envuelve y comparándola con regiones de los casos de ejemplo de la biblioteca. Esto era ineficiente así que se han introducido mejoras que van desde la carga y preselección de los casos más semejantes para usarlos en la segmentación hasta introducir una estimación pre calculada del etiquetado de los voxels que no suelen variar para evitar tener que procesarlos. Resultados: Se parte de un método que obtiene segmentaciones con una 98% de fiabilidad y en un tiempo de ejecución de 160 segundos y se ha mejorado hasta una fiabilidad del 99% en un tiempo inferior a 40 segundos.