Autism spectrum disorder (ASD) is a complex neuro-developmental disorder that affects social skills, language, speech and communication. Early detection of ASD individuals, especially children, could help to devise and strategize right therapeutic plan at right time. Human faces encode important markers that can be used to identify ASD by analyzing facial features, eye contact, and so on. In this work, an improved transfer-learning-based autism face recognition framework is proposed to identify kids with ASD in the early stages more precisely. Therefore, we have collected face images of children with ASD from the Kaggle data repository, and various machine learning and deep learning classifiers and other transfer-learning-based pre-trained models were applied. We observed that our improved MobileNet-V1 model demonstrates the best accuracy of 90.67% and the lowest 9.33% value of both fall-out and miss rate compared to the other classifiers and pre-trained models. Furthermore, this classifier is used to identify different ASD groups investigating only autism image data using k-means clustering technique. Thus, the improved MobileNet-V1 model showed the highest accuracy (92.10%) for k = 2 autism sub-types. We hope this model will be useful for physicians to detect autistic children more explicitly at the early stage.

We have previously shown that long-term administration of aminoguanidine, an inhibitor of advanced glycosylation product formation, reduces the extent of experimental diabetic retinopathy in the rat by 85%. In order to determine whether the residual retinopathy that developed despite aminoguanidine was attributable to advanced glycation endproduct formation, a time-course study was performed in three different groups of male Wistar rats: non-diabetic controls (NC), streptozotocin-diabetic controls (DC) and streptozotocin-diabetic rats treated with aminoguanidine HCL, 50 mg/100 ml drinking water (D-AG). Eyes were obtained at 24, 32, 44 and 56 weeks of diabetes/treatment duration and morphologic evaluation was done on retinal digest preparations. At 56 weeks, retinal basement membrane thickness was additionally measured. After 24 weeks of diabetes, the number of acellular capillaries was significantly elevated in DC (44.6 +/- 5.7/mm2 of retinal area, NC 19.6 +/- 4.9; p < 0.001) and increased continuously over time (DC 56 weeks 87.4 +/- 15.1; p < 0.001 vs DC24 weeks). In contrast, acellular capillaries in D-AG increased over the first 24 weeks and then remained constant for the rest of the study (D-AG 24 weeks 35.7 +/- 5.18; p < 0.01 vs NC 24 weeks and NS vs DC 24 weeks; D-AG 56 weeks 42.0 +/- 6.20; p NS vs D-AG 24 weeks).(ABSTRACT TRUNCATED AT 250 WORDS)

No abstract

The endocrine pancreas of four reptile species belonging to the turtles, lizards and snakes was investigated immunohistochemically for the occurrence and cellular distribution of chromogranin A (CgA) and of two synthetic secretonin II (SgII)-peptides (C23-3 and C26-3). CgA-immunoreactivity was found only in the turtle pancreas, whereas that for SGIIC23-3 appeared both in the turtle and snake. None of the species studied displayed immunoreactivity for SgIIC26-3. The two detected granins showed different distributions in relation to the endocrine cell types. Conspicuous variations of the immunostaining density for either granin in the same endocrine cell population and even complete lack of the immunoreaction were recorded. The findings suggest that, despite the restricted presence in the endocrine pancreas of the reptiles investigated, the granins are relatively well conserved during phylogeny; they do not confirm, however, the previously accepted usefulness of the granin protein family as common markers of neuroendocrine cells.

In the pancreas of Scyliorhinus stellaris large islets are usually found around small ducts, the inner surface of which is covered by elongated epithelial cells; thus the endocrine cells are never exposed directly to the lumen of the duct. Sometimes, single islet cells or small groups of endocrine elements are also incorporated into acini. Using correlative light and electron microscopy, eight islet cell types were identified: Only B-cells (type I) display a positive reaction with pseudoisocyanin and aldehyde-fuchsin staining. This cell type contains numerous small secretory granules (diameter 280 nm). Type II- and III-cells possess large granules stainable with orange G and azocarmine and show strong luminescence with dark-field microscopy. Type II-cells have spherical (diameter 700 nm), type III-cells spherical to elongated granules (diameter 450 x 750 nm). Type II-cells are possibly analogous to A-cells, while type III-cells resemble mammalian enterochromaffin cells. Type IV-cells contain granules (diameter 540 nm) of high electron density showing a positive reaction to the Hellman-Hellerström silver impregnation and a negative reaction to Grimelius' silver impregnation; they are most probably analogous to D-cells of other species. Type V-cells exhibit smaller granules (diameter 250 x 500 nm), oval to elongated in shape. Type VI-cells contain small spherical granules (diameter 310 nm). Type VII-cells possess two kinds of large granules interspersed in the cytoplasm; one type is spherical and electron dense (diameter 650 nm), the other spherical and less electron dense (diameter 900 nm). Type VIII-cells have small granules curved in shape and show moderate electron density (diameter 100 nm). Grimelius-positive secretory granules were not only found in cell types II and III, but also in types V, VI, and VII. B-cells (type I) and the cell types II to IV were the most frequent cells; types V to VII occurred occasionally, whereas type VIII-cells were very rare.