Effective diagnosis of Alzheimer's disease by means of large margin-based methodology.

BACKGROUND Functional brain images such as Single-Photon Emission Computed Tomography (SPECT) and Positron Emission Tomography (PET) have been widely used to guide the clinicians in the Alzheimer's Disease (AD) diagnosis. However, the subjectivity involved in their evaluation has favoured the development of Computer Aided Diagnosis (CAD) Systems. METHODS It is proposed a novel combination of feature extraction techniques to improve the diagnosis of AD. Firstly, Regions of Interest (ROIs) are selected by means of a t-test carried out on 3D Normalised Mean Square Error (NMSE) features restricted to be located within a predefined brain activation mask. In order to address the small sample-size problem, the dimension of the feature space was further reduced by: Large Margin Nearest Neighbours using a rectangular matrix (LMNN-RECT), Principal Component Analysis (PCA) or Partial Least Squares (PLS) (the two latter also analysed with a LMNN transformation). Regarding the classifiers, kernel Support Vector Machines (SVMs) and LMNN using Euclidean, Mahalanobis and Energy-based metrics were compared. RESULTS Several experiments were conducted in order to evaluate the proposed LMNN-based feature extraction algorithms and its benefits as: i) linear transformation of the PLS or PCA reduced data, ii) feature reduction technique, and iii) classifier (with Euclidean, Mahalanobis or Energy-based methodology). The system was evaluated by means of k-fold cross-validation yielding accuracy, sensitivity and specificity values of 92.78%, 91.07% and 95.12% (for SPECT) and 90.67%, 88% and 93.33% (for PET), respectively, when a NMSE-PLS-LMNN feature extraction method was used in combination with a SVM classifier, thus outperforming recently reported baseline methods. CONCLUSIONS All the proposed methods turned out to be a valid solution for the presented problem. One of the advances is the robustness of the LMNN algorithm that not only provides higher separation rate between the classes but it also makes (in combination with NMSE and PLS) this rate variation more stable. In addition, their generalization ability is another advance since several experiments were performed on two image modalities (SPECT and PET).

Fine-tuning the space, time, and host distribution of mycobacteria in wildlife.

Background. We describe the diversity of two kinds of mycobacteria isolates, environmental mycobacteria and Mycobacterium bovis collected from wild boar, fallow deer, red deer and cattle in Doñana National Park (DNP, Spain), analyzing their association with temporal, spatial and environmental factors. Results. High diversity of environmental mycobacteria species and M. bovis typing patterns (TPs) were found. When assessing the factors underlying the presence of the most common types of both environmental mycobacteria and M. bovis TPs in DNP, we evidenced (i) host species differences in the occurrence, (ii) spatial structuration and (iii) differences in the degree of spatial association of specific types between host species. Co-infection of a single host by two M. bovis TPs occurred in all three wild ungulate species. In wild boar and red deer, isolation of one group of mycobacteria occurred more frequently in individuals not infected by the other group. While only three TPs were detected in wildlife between 1998 and 2003, up to 8 different ones were found during 2006-2007. The opposite was observed in cattle. Belonging to an M. bovis-infected social group was a significant risk factor for mycobacterial infection in red deer and wild boar, but not for fallow deer. M. bovis TPs were usually found closer to water marshland than MOTT. Conclusions. The diversity of mycobacteria described herein is indicative of multiple introduction events and a complex multi-host and multi-pathogen epidemiology in DNP. Significant changes in the mycobacterial isolate community may have taken place, even in a short time period (1998 to 2007). Aspects of host social organization should be taken into account in wildlife epidemiology. Wildlife in DNP is frequently exposed to different species of non-tuberculous, environmental mycobacteria, which could interact with the immune response to pathogenic mycobacteria, although the effects are unknown. This research highlights the suitability of molecular typing for surveys at small spatial and temporal scales.