Search

Objectives:A large retrospective multicentre study was conducted in Spain to evaluate the efficiency of the new European Society for Pediatric Gastroenterology, Hepatology, and Nutrition (ESPGHAN) criteria for the diagnosis of coeliac disease (CD).Methods:The study protocol was approved by the ethics committee of Hospital Universitari i Politecnic La Fe (Valencia, Spain). The present study included 2177 children (ages 0.6-15.9 years) with small bowel biopsy (SBB) performed for diagnostic purposes (from 2000 to 2009) and with a minimum 2-year follow-up after biopsy.Results:CD was diagnosed in 2126 patients (97.5%) and excluded in 51 (2.5%). Tissue transglutaminase antibodies (TG2A), anti-endomysial antibodies (EMA), and human leukocyte antigen (HLA) were reported in 751 patients, 640 symptomatic and 111 asymptomatic. TG2A levels >10 times the upper limit of normal, plus positive EMA and HLA DQ2 and/or DQ8 haplotypes, were found in 336 symptomatic patients, all of them with final diagnosis of CD. In 65 of 69 asymptomatic patients, 65 had confirmed CD and 4 did not have CD. According to the 2012 ESPGHAN guidelines, SBB may have been omitted in 52% of the symptomatic patients with CD with serologic and HLA available data. Gluten challenge was performed in 158 children, 75 of them

Gender classification (GC) has achieved high accuracy in different experimental evaluations based mostly on inner facial details. However, these results do not generalize well in unrestricted datasets and particularly in cross-database experiments, where the performance drops drastically. In this paper, we analyze the state-of-the-art GC accuracy on three large datasets: MORPH, LFW and GROUPS. We discuss their respective difficulties and bias, concluding that the most challenging and wildest complexity is present in GROUPS. This dataset covers hard conditions such as low resolution imagery and cluttered background. Firstly, we analyze in depth the performance of different descriptors extracted from the face and its local context on this dataset. Selecting the bests and studying their most suitable combination allows us to design a solution that beats any previously published results for GROUPS with the Dago's protocol, reaching an accuracy over 94.2%, reducing the gap with other simpler datasets. The chosen solution based on local descriptors is later evaluated in a cross-database scenario with the three mentioned datasets, and full dataset 5-fold cross validation. The achieved results are compared with a Convolutional Neural Network approach, achieving rather similar marks. Finally, a solution is proposed combining both focuses, exhibiting great complementarity, boosting GC performance to beat previously published results in GC both cross-database, and full in-database evaluations., Revised version containing 12 pages. This revision includes newer referenes, results with CNN, fusion of local descriptors amd CNN and corrects different typos