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).

Efectividad de una estrategia de intervención preventiva, basada en entrevistas telefónicas estructuradas, en una población laboral con riesgo cardiovascular moderado/alto.

OBJECTIVE This study assesses the effectiveness of a structured telephone survey on cardiovascular prevention, in modifying lifestyle, on cardiovascular risk parameters, percentage of smoking cessation and overall cardiovascular risk (CVR). DESIGN Quasi-experimental study of preventive intervention. SETTING Ibermutuamur (Spanish Accident and Health Insurance Company). Centres established throughout Spain. PARTICIPANTS A total of 4,792 workers with moderate/high cardiovascular risk who had agreed to be contacted by phone. Subjects with a previous diagnosis of cardiovascular disease and those receiving treatment for hypertension, hypercholesterolemia or diabetes were excluded. INTERVENTION A final total of 3,085 workers were contacted and were followed up by telephone surveys on the first, fourth and eighth month after the initial check up (CU) in order to emphasise cardiovascular health advice (Group A); we failed to contact 1,707 workers, who only attended the baseline and one year CUs (Group B). PRINCIPAL OUTCOMES: CUs included medical records and physical examination, with two blood pressure measurements, Body Mass Index (BMI), and biochemical parameters. Cardiovascular risk was stratified following the European cardiovascular SCORE. Individuals with a relative risk higher than 4 were also considered as high-risk. All workers were informed about their cardiovascular risk profile (CVRF) and healthy cardiovascular lifestyle measures. They were also given a letter for their General Practitioner (GP) to inform them on the worker's cardiovascular risk level. RESULTS A total of 71.5% of the workers were over 45 years, 95.0% males, 76.6% manual workers ("Blue Collar") and 69.7% smokers. Both groups showed improvement in lipid parameters, blood pressure, smoking cessation and overall cardiovascular risk in the second CU. There were significant differences in favour of Group A as regards blood pressure, lipids (except HDL cholesterol), BMI, glycaemia, smoking cessation (A: 23.5%/B: 19.44%, P=0.001) and CVR stratum improvement (A: 46.6%/B: 37.7%, P=0.0001). The large majority (85%) of workers read preventive recommendations; 33% knew their risk level and 73% knew their CVRF. 52.9% gave the letter to the GP, which led them to start therapies on diet (47%), hypertension (19.5%), dyslipidaemia (16.7%), diabetes (4.4%) and smoking (2.9%) and no changes were made in 36.5% of cases. CONCLUSIONS The results of this study suggests that cardiovascular prevention strategy based on structured telephone surveys on high/moderate CVR subjects to promote lifestyle changes could be effective at reducing CVR. A clinical trial is required for confirmation. Sending information on CVRF following routine medial CUs and Primary Care involvement, could contribute to the positive changes observed.