Segmentação do tecido tumoral cerebral em imagens de ressonância magnética

Dissertação de mestrado integrado em Engenharia Biomédica (área de especialização em Eletrónica Médica)

Does introduction of thresholds in decision aids benefit the patient?: Comparison between findings-based and threshold-based diagnostic decision aids.

PURPOSE: To assess how different diagnostic decision aids perform in terms of sensitivity, specificity, and harm. METHODS: Four diagnostic decision aids were compared, as applied to a simulated patient population: a findings-based algorithm following a linear or branched pathway, a serial threshold-based strategy, and a parallel threshold-based strategy. Headache in immune-compromised HIV patients in a developing country was used as an example. Diagnoses included cryptococcal meningitis, cerebral toxoplasmosis, tuberculous meningitis, bacterial meningitis, and malaria. Data were derived from literature and expert opinion. Diagnostic strategies' validity was assessed in terms of sensitivity, specificity, and harm related to mortality and morbidity. Sensitivity analyses and Monte Carlo simulation were performed. RESULTS: The parallel threshold-based approach led to a sensitivity of 92% and a specificity of 65%. Sensitivities of the serial threshold-based approach and the branched and linear algorithms were 47%, 47%, and 74%, respectively, and the specificities were 85%, 95%, and 96%. The parallel threshold-based approach resulted in the least harm, with the serial threshold-based approach, the branched algorithm, and the linear algorithm being associated with 1.56-, 1.44-, and 1.17-times higher harm, respectively. Findings were corroborated by sensitivity and Monte Carlo analyses. CONCLUSION: A threshold-based diagnostic approach is designed to find the optimal trade-off that minimizes expected harm, enhancing sensitivity and lowering specificity when appropriate, as in the given example of a symptom pointing to several life-threatening diseases. Findings-based algorithms, in contrast, solely consider clinical observations. A parallel workup, as opposed to a serial workup, additionally allows for all potential diseases to be reviewed, further reducing false negatives. The parallel threshold-based approach might, however, not be as good in other disease settings.

Quantitative morphometry analysis of the fetal brain using clinical MR imaging

In vivo fetal magnetic resonance imaging provides aunique approach for the study of early human braindevelopment [1]. In utero cerebral morphometry couldpotentially be used as a marker of the cerebralmaturation and help to distinguish between normal andabnormal development in ambiguous situations. However,this quantitative approach is a major challenge becauseof the movement of the fetus inside the amniotic cavity,the poor spatial resolution provided by very fast MRIsequences and the partial volume effect. Extensiveefforts are made to deal with the reconstruction ofhigh-resolution 3D fetal volumes based on severalacquisitions with lower resolution [2,3,4]. Frameworkswere developed for the segmentation of specific regionsof the fetal brain such as posterior fossa, brainstem orgerminal matrix [5,6], or for the entire brain tissue[7,8], applying the Expectation-Maximization MarkovRandom Field (EM-MRF) framework. However, many of theseprevious works focused on the young fetus (i.e. before 24weeks) and use anatomical atlas priors to segment thedifferent tissue or regions. As most of the gyraldevelopment takes place after the 24th week, acomprehensive and clinically meaningful study of thefetal brain should not dismiss the third trimester ofgestation. To cope with the rapidly changing appearanceof the developing brain, some authors proposed a dynamicatlas [8]. To our opinion, this approach however faces arisk of circularity: each brain will be analyzed /deformed using the template of its biological age,potentially biasing the effective developmental delay.Here, we expand our previous work [9] to proposepost-processing pipeline without prior that allow acomprehensive set of morphometric measurement devoted toclinical application. Data set & Methods: Prenatal MRimaging was performed with a 1-T system (GE MedicalSystems, Milwaukee) using single shot fast spin echo(ssFSE) sequences (TR 7000 ms, TE 180 ms, FOV 40 x 40 cm,slice thickness 5.4mm, in plane spatial resolution1.09mm). For each fetus, 6 axial volumes shifted by 1 mmwere acquired under motherâeuro?s sedation (about 1min pervolume). First, each volume is segmentedsemi-automatically using region-growing algorithms toextract fetal brain from surrounding maternal tissues.Inhomogeneity intensity correction [10] and linearintensity normalization are then performed. Brain tissues(CSF, GM and WM) are then segmented based on thelow-resolution volumes as presented in [9]. Ahigh-resolution image with isotropic voxel size of 1.09mm is created as proposed in [2] and using B-splines forthe scattered data interpolation [11]. Basal gangliasegmentation is performed using a levet setimplementation on the high-resolution volume [12]. Theresulting white matter image is then binarized and givenas an input in FreeSurfer software(http://surfer.nmr.mgh.harvard.edu) to providetopologically accurate three-dimensional reconstructionsof the fetal brain according to the local intensitygradient. References: [1] Guibaud, Prenatal Diagnosis29(4) (2009). [2] Rousseau, Acad. Rad. 13(9), 2006. [3]Jiang, IEEE TMI 2007. [4] Warfield IADB, MICCAI 2009. [5]Claude, IEEE Trans. Bio. Eng. 51(4) 2004. [6] Habas,MICCAI 2008. [7] Bertelsen, ISMRM 2009. [8] Habas,Neuroimage 53(2) 2010. [9] Bach Cuadra, IADB, MICCAI2009. [10] Styner, IEEE TMI 19(39 (2000). [11] Lee, IEEETrans. Visual. And Comp. Graph. 3(3), 1997. [12] BachCuadra, ISMRM 2010.

Learning Preferences with Kernel-Based Methods

Learning of preference relations has recently received significant attention in machine learning community. It is closely related to the classification and regression analysis and can be reduced to these tasks. However, preference learning involves prediction of ordering of the data points rather than prediction of a single numerical value as in case of regression or a class label as in case of classification. Therefore, studying preference relations within a separate framework facilitates not only better theoretical understanding of the problem, but also motivates development of the efficient algorithms for the task. Preference learning has many applications in domains such as information retrieval, bioinformatics, natural language processing, etc. For example, algorithms that learn to rank are frequently used in search engines for ordering documents retrieved by the query. Preference learning methods have been also applied to collaborative filtering problems for predicting individual customer choices from the vast amount of user generated feedback. In this thesis we propose several algorithms for learning preference relations. These algorithms stem from well founded and robust class of regularized least-squares methods and have many attractive computational properties. In order to improve the performance of our methods, we introduce several non-linear kernel functions. Thus, contribution of this thesis is twofold: kernel functions for structured data that are used to take advantage of various non-vectorial data representations and the preference learning algorithms that are suitable for different tasks, namely efficient learning of preference relations, learning with large amount of training data, and semi-supervised preference learning. Proposed kernel-based algorithms and kernels are applied to the parse ranking task in natural language processing, document ranking in information retrieval, and remote homology detection in bioinformatics domain. Training of kernel-based ranking algorithms can be infeasible when the size of the training set is large. This problem is addressed by proposing a preference learning algorithm whose computation complexity scales linearly with the number of training data points. We also introduce sparse approximation of the algorithm that can be efficiently trained with large amount of data. For situations when small amount of labeled data but a large amount of unlabeled data is available, we propose a co-regularized preference learning algorithm. To conclude, the methods presented in this thesis address not only the problem of the efficient training of the algorithms but also fast regularization parameter selection, multiple output prediction, and cross-validation. Furthermore, proposed algorithms lead to notably better performance in many preference learning tasks considered.

Fourier-based registration for robust forward-looking sonar mosaicing in low-visibility underwater environments

Vehicle operations in underwater environments are often compromised by poor visibility conditions. For instance, the perception range of optical devices is heavily constrained in turbid waters, thus complicating navigation and mapping tasks in environments such as harbors, bays, or rivers. A new generation of high-definition forward-looking sonars providing acoustic imagery at high frame rates has recently emerged as a promising alternative for working under these challenging conditions. However, the characteristics of the sonar data introduce difficulties in image registration, a key step in mosaicing and motion estimation applications. In this work, we propose the use of a Fourier-based registration technique capable of handling the low resolution, noise, and artifacts associated with sonar image formation. When compared to a state-of-the art region-based technique, our approach shows superior performance in the alignment of both consecutive and nonconsecutive views as well as higher robustness in featureless environments. The method is used to compute pose constraints between sonar frames that, integrated inside a global alignment framework, enable the rendering of consistent acoustic mosaics with high detail and increased resolution. An extensive experimental section is reported showing results in relevant field applications, such as ship hull inspection and harbor mapping

Recent advances in monocular model-based tracking: a systematic literature review

In this paper, we review the advances of monocular model-based tracking for last ten years period until 2014. In 2005, Lepetit, et. al, [19] reviewed the status of monocular model based rigid body tracking. Since then, direct 3D tracking has become quite popular research area, but monocular model-based tracking should still not be forgotten. We mainly focus on tracking, which could be applied to aug- mented reality, but also some other applications are covered. Given the wide subject area this paper tries to give a broad view on the research that has been conducted, giving the reader an introduction to the different disciplines that are tightly related to model-based tracking. The work has been conducted by searching through well known academic search databases in a systematic manner, and by selecting certain publications for closer examination. We analyze the results by dividing the found papers into different categories by their way of implementation. The issues which have not yet been solved are discussed. We also discuss on emerging model-based methods such as fusing different types of features and region-based pose estimation which could show the way for future research in this subject.

Region Connection Calculus: Composition Tables and Constraint Satisfaction Problems

Qualitative spatial reasoning (QSR) is an important field of AI that deals with qualitative aspects of spatial entities. Regions and their relationships are described in qualitative terms instead of numerical values. This approach models human based reasoning about such entities closer than other approaches. Any relationships between regions that we encounter in our daily life situations are normally formulated in natural language. For example, one can outline one's room plan to an expert by indicating which rooms should be connected to each other. Mereotopology as an area of QSR combines mereology, topology and algebraic methods. As mereotopology plays an important role in region based theories of space, our focus is on one of the most widely referenced formalisms for QSR, the region connection calculus (RCC). RCC is a first order theory based on a primitive connectedness relation, which is a binary symmetric relation satisfying some additional properties. By using this relation we can define a set of basic binary relations which have the property of being jointly exhaustive and pairwise disjoint (JEPD), which means that between any two spatial entities exactly one of the basic relations hold. Basic reasoning can now be done by using the composition operation on relations whose results are stored in a composition table. Relation algebras (RAs) have become a main entity for spatial reasoning in the area of QSR. These algebras are based on equational reasoning which can be used to derive further relations between regions in a certain situation. Any of those algebras describe the relation between regions up to a certain degree of detail. In this thesis we will use the method of splitting atoms in a RA in order to reproduce known algebras such as RCC15 and RCC25 systematically and to generate new algebras, and hence a more detailed description of regions, beyond RCC25.

Bee Communities in Restored Landfill Sites of Niagara Region

This study examined the impact of habitat restoration on bee communities (Hymenoptera: Apidae) of the Niagara Region, Ontario, Canada. Bee abundance and diversity was studied in three restored landfill sites: the Glenridge Quarry Naturalization Site (GQNS) in St. Catharines, Elm Street Naturalization Site in Port Colborne, and Station Road Naturalization Site in Wainfleet during 2011 and 2012. GQNS represented older sites restored from 2001-2003. Elm and Station sites represented newly restored landfills as of 2011. These sites were compared to control sites at Brock University where bee communities are well established and again to other landfills where no stable habitat was available before restoration. The objective of this study is to investigate the impact of restoration level on bee abundance and diversity in restored landfill sites of the Niagara Region. Based on the increased disturbance hypothesis (InDH) and the intermediate disturbance hypothesis (IDH), I hypothesized that bee abundance and diversity will follow two patterns. First pattern according to InDH suggest that as the disturbance decrease the bee abundance and diversity will increased. Second pattern according to the IDH bee abundance and diversity will be the highest at the intermediate level of disturbance. A total of 7 173 bees were collected using pan traps and flower collections, from May to October 2011 and 2012. Bees were classified to five families, 21 genera and sub-genera, containing at least 78 species. In 2011 bee abundance was not significantly different among restoration levels while in 2012 bee abundance was significant difference among restoration level. According to family there were no significant difference in Halictidae and Apidae abundance among restoration level while Colletidae and Megachilidae abundance were varied among restoration levels. The bee species richness was highest in the newly restored sites followed by restored control sites, and then the control site. The current study demonstrates that habitat restoration results in rapid increases in bee abundance and diversity for newly restored sites, and, further, that it takes only 2-3 years for bee assemblages in newly restored sites to arrive at the same levels of abundance and diversity as in nearby control sites where bee communities are well established.

The use of artificial intelligence algorithms to guide surgical treatment of adolescent idiopathic scoliosis

La scoliose idiopathique de l’adolescent (SIA) est une déformation tri-dimensionelle du rachis. Son traitement comprend l’observation, l’utilisation de corsets pour limiter sa progression ou la chirurgie pour corriger la déformation squelettique et cesser sa progression. Le traitement chirurgical reste controversé au niveau des indications, mais aussi de la chirurgie à entreprendre. Malgré la présence de classifications pour guider le traitement de la SIA, une variabilité dans la stratégie opératoire intra et inter-observateur a été décrite dans la littérature. Cette variabilité s’accentue d’autant plus avec l’évolution des techniques chirurgicales et de l’instrumentation disponible. L’avancement de la technologie et son intégration dans le milieu médical a mené à l’utilisation d’algorithmes d’intelligence artificielle informatiques pour aider la classification et l’évaluation tridimensionnelle de la scoliose. Certains algorithmes ont démontré être efficace pour diminuer la variabilité dans la classification de la scoliose et pour guider le traitement. L’objectif général de cette thèse est de développer une application utilisant des outils d’intelligence artificielle pour intégrer les données d’un nouveau patient et les évidences disponibles dans la littérature pour guider le traitement chirurgical de la SIA. Pour cela une revue de la littérature sur les applications existantes dans l’évaluation de la SIA fut entreprise pour rassembler les éléments qui permettraient la mise en place d’une application efficace et acceptée dans le milieu clinique. Cette revue de la littérature nous a permis de réaliser que l’existence de “black box” dans les applications développées est une limitation pour l’intégration clinique ou la justification basée sur les évidence est essentielle. Dans une première étude nous avons développé un arbre décisionnel de classification de la scoliose idiopathique basé sur la classification de Lenke qui est la plus communément utilisée de nos jours mais a été critiquée pour sa complexité et la variabilité inter et intra-observateur. Cet arbre décisionnel a démontré qu’il permet d’augmenter la précision de classification proportionnellement au temps passé à classifier et ce indépendamment du niveau de connaissance sur la SIA. Dans une deuxième étude, un algorithme de stratégies chirurgicales basé sur des règles extraites de la littérature a été développé pour guider les chirurgiens dans la sélection de l’approche et les niveaux de fusion pour la SIA. Lorsque cet algorithme est appliqué à une large base de donnée de 1556 cas de SIA, il est capable de proposer une stratégie opératoire similaire à celle d’un chirurgien expert dans prêt de 70% des cas. Cette étude a confirmé la possibilité d’extraire des stratégies opératoires valides à l’aide d’un arbre décisionnel utilisant des règles extraites de la littérature. Dans une troisième étude, la classification de 1776 patients avec la SIA à l’aide d’une carte de Kohonen, un type de réseaux de neurone a permis de démontrer qu’il existe des scoliose typiques (scoliose à courbes uniques ou double thoracique) pour lesquelles la variabilité dans le traitement chirurgical varie peu des recommandations par la classification de Lenke tandis que les scolioses a courbes multiples ou tangentielles à deux groupes de courbes typiques étaient celles avec le plus de variation dans la stratégie opératoire. Finalement, une plateforme logicielle a été développée intégrant chacune des études ci-dessus. Cette interface logicielle permet l’entrée de données radiologiques pour un patient scoliotique, classifie la SIA à l’aide de l’arbre décisionnel de classification et suggère une approche chirurgicale basée sur l’arbre décisionnel de stratégies opératoires. Une analyse de la correction post-opératoire obtenue démontre une tendance, bien que non-statistiquement significative, à une meilleure balance chez les patients opérés suivant la stratégie recommandée par la plateforme logicielle que ceux aillant un traitement différent. Les études exposées dans cette thèse soulignent que l’utilisation d’algorithmes d’intelligence artificielle dans la classification et l’élaboration de stratégies opératoires de la SIA peuvent être intégrées dans une plateforme logicielle et pourraient assister les chirurgiens dans leur planification préopératoire.

Development of Shape Descriptors Based on Legendre Polynomials and Content-Based Retrieval of Scoliosis Images

The wealth of information available freely on the web and medical image databases poses a major problem for the end users: how to find the information needed? Content –Based Image Retrieval is the obvious solution.A standard called MPEG-7 was evolved to address the interoperability issues of content-based search.The work presented in this thesis mainly concentrates on developing new shape descriptors and a framework for content – based retrieval of scoliosis images.New region-based and contour based shape descriptor is developed based on orthogonal Legendre polymomials.A novel system for indexing and retrieval of digital spine radiographs with scoliosis is presented.

Content Based Image Retrieval Using Low Level Features of Automatically Extracted Regions of Interest

This paper proposes a region based image retrieval system using the local colour and texture features of image sub regions. The regions of interest (ROI) are roughly identified by segmenting the image into fixed partitions, finding the edge map and applying morphological dilation. The colour and texture features of the ROIs are computed from the histograms of the quantized HSV colour space and Gray Level co- occurrence matrix (GLCM) respectively. Each ROI of the query image is compared with same number of ROIs of the target image that are arranged in the descending order of white pixel density in the regions, using Euclidean distance measure for similarity computation. Preliminary experimental results show that the proposed method provides better retrieving result than retrieval using some of the existing methods.

Content-Based Image Retrieval of Axial Brain Slices Using a Novel LBP with a Ternary Encoding

Retrieval of similar anatomical structures of brain MR images across patients would help the expert in diagnosis of diseases. In this paper, modified local binary pattern with ternary encoding called modified local ternary pattern (MOD-LTP) is introduced, which is more discriminant and less sensitive to noise in near-uniform regions, to locate slices belonging to the same level from the brain MR image database. The ternary encoding depends on a threshold, which is a user-specified one or calculated locally, based on the variance of the pixel intensities in each window. The variancebased local threshold makes the MOD-LTP more robust to noise and global illumination changes. The retrieval performance is shown to improve by taking region-based moment features of MODLTP and iteratively reweighting the moment features of MOD-LTP based on the user’s feedback. The average rank obtained using iterated and weighted moment features of MOD-LTP with a local variance-based threshold, is one to two times better than rotational invariant LBP (Unay, D., Ekin, A. and Jasinschi, R.S. (2010) Local structure-based region-of-interest retrieval in brain MR images. IEEE Trans. Inf. Technol. Biomed., 14, 897–903.) in retrieving the first 10 relevant images

Investigation into the performance of an algorithm for deriving concise descriptions of optic flow fields for real-time control

This paper describes a region-based algorithm for deriving a concise description of a first order optical flow field. The algorithm described achieves performance improvements over existing algorithms without compromising the accuracy of the flow field values calculated. These improvements are brought about by not computing the entire flow field between two consecutive images, but by considering only the flow vectors of a selected subset of the images. The algorithm is presented in the context of a project to balance a bipedal robot using visual information.

A comparison between feature-based and EM-based contour tracking

Most active-contour methods are based either on maximizing the image contrast under the contour or on minimizing the sum of squared distances between contour and image 'features'. The Marginalized Likelihood Ratio (MLR) contour model uses a contrast-based measure of goodness-of-fit for the contour and thus falls into the first class. The point of departure from previous models consists in marginalizing this contrast measure over unmodelled shape variations. The MLR model naturally leads to the EM Contour algorithm, in which pose optimization is carried out by iterated least-squares, as in feature-based contour methods. The difference with respect to other feature-based algorithms is that the EM Contour algorithm minimizes squared distances from Bayes least-squares (marginalized) estimates of contour locations, rather than from 'strongest features' in the neighborhood of the contour. Within the framework of the MLR model, alternatives to the EM algorithm can also be derived: one of these alternatives is the empirical-information method. Tracking experiments demonstrate the robustness of pose estimates given by the MLR model, and support the theoretical expectation that the EM Contour algorithm is more robust than either feature-based methods or the empirical-information method. (c) 2005 Elsevier B.V. All rights reserved.

Evaluation of satellite-based and model re-analysis rainfall estimates for Uganda

The dependence of much of Africa on rain fed agriculture leads to a high vulnerability to fluctuations in rainfall amount. Hence, accurate monitoring of near-real time rainfall is particularly useful, for example in forewarning possible crop shortfalls in drought-prone areas. Unfortunately, ground based observations are often inadequate. Rainfall estimates from satellite-based algorithms and numerical model outputs can fill this data gap, however rigorous assessment of such estimates is required. In this case, three satellite based products (NOAA-RFE 2.0, GPCP-1DD and TAMSAT) and two numerical model outputs (ERA-40 and ERA-Interim) have been evaluated for Uganda in East Africa using a network of 27 rain gauges. The study focuses on the years 2001 to 2005 and considers the main rainy season (February to June). All data sets were converted to the same temporal and spatial scales. Kriging was used for the spatial interpolation of the gauge data. All three satellite products showed similar characteristics and had a high level of skill that exceeded both model outputs. ERA-Interim had a tendency to overestimate whilst ERA-40 consistently underestimated the Ugandan rainfall.