Automatic severity grade classification of diabetic retinopathy using deformable ladder Bi attention U-net and deep adaptive CNN.

Long-term exposure to diabetes mellitus leads to the formation of diabetic retinopathy (DR), which can cause vision loss in working-age adults. Early stage diagnosis of DR is highly essential for preventing vision loss and preserving vision in people with diabetes. The motivation behind the severity grade classification of DR is to develop an automated system that can assist ophthalmologists and healthcare professionals in the diagnosis and management of DR. However, existing methods suffer from variability in image quality, similar structures of the normal and lesion regions, high dimensional features, variability in disease manifestations, small datasets, high training loss, model complexity, and overfitting, which leads to high misclassification errors in the severity grading system. Hence, there is a need to develop an automated system using improved deep learning techniques to provide a reliable and consistent grading of DR severity with high classification accuracy using fundus images. To solve these issues, we proposes a Deformable Ladder Bi attention U-shaped encoder-decoder network and Deep Adaptive Convolutional Neural Network (DLBUnet-DACNN) for accurate severity classification of DR. The DLBUnet performs lesion segmentation that can be divided into three parts: the encoder, the central processing module and the decoder. In the encoder part, deformable convolution is used instead of convolution to learn different shapes of the lesion by understanding the offset location. Afterwards, Ladder Atrous Spatial Pyramidal Pooling (LASPP) using variable dilation rates is introduced in the central processing module. LASPP enhance the tiny lesion features and variable dilation rates avoid gridding effects and can learn better global context information. Then the decoder part uses a bi-attention layer contains spatial and channel attention, which can learn contour and edges of the lesion accurately. Finally, the severity of DR is classified using a DACNN by extracting the discriminative features from the segmentation results. Experiments are conducted on the Messidor-2, Kaggle, and Messidor datasets. Our proposed method DLBUnet-DACNN achieves better results in terms of accuracy of 98.2, recall of 0.987, kappa coefficient of 0.993, precision of 0.98, F1-score of 0.981, Matthews Correlation Coefficient (MCC) of 0.93 and Classification Success Index (CSI) of 0.96 when compared to existing methods.

Detection of Hard Exudates in Colour Fundus Images Using Fuzzy Support Vector Machine-Based Expert System.

Diabetic retinopathy is a major cause of vision loss in diabetic patients. Currently, there is a need for making decisions using intelligent computer algorithms when screening a large volume of data. This paper presents an expert decision-making system designed using a fuzzy support vector machine (FSVM) classifier to detect hard exudates in fundus images. The optic discs in the colour fundus images are segmented to avoid false alarms using morphological operations and based on circular Hough transform. To discriminate between the exudates and the non-exudates pixels, colour and texture features are extracted from the images. These features are given as input to the FSVM classifier. The classifier analysed 200 retinal images collected from diabetic retinopathy screening programmes. The tests made on the retinal images show that the proposed detection system has better discriminating power than the conventional support vector machine. With the best combination of FSVM and features sets, the area under the receiver operating characteristic curve reached 0.9606, which corresponds to a sensitivity of 94.1% with a specificity of 90.0%. The results suggest that detecting hard exudates using FSVM contribute to computer-assisted detection of diabetic retinopathy and as a decision support system for ophthalmologists.

Effect of dietary fat intake and exercise on inflammatory mediators of the immune system in sedentary men and women.

OBJECTIVE: Dietary fat intake and exercise affect the immune system. This study determined the changes in inflammatory components of the immune system in response to maximal exercise with three levels of dietary fat intake: 19%, 30%, and 50% of total calories. METHODS: Five men and six women were randomly assigned to consume diets with 19% and 50% calories from fat for three weeks each, with a one-week washout. The habitual and washout diets were 30% calories from fat. At the beginning and the end of each diet, body composition and maximal exercise tests were performed. Blood samples were collected before and after exercise to determine the immunological parameters. RESULTS: The subject's energy intake was balanced to expenditure on the 30% and 50% diets, but was in negative balance on the 19% diet. Exercise led to significant increases in the concentrations of leukocytes, neutrophils, lymphocytes, monocytes, plasma tumour necrosis factor (TNF)-alpha, plasma interleukin (IL)-2, plasma soluble vascular cell adhesion molecule (sVCAM)-1, and the production of IL-1beta and IL-6 by peripheral blood mononuclear (PBMN) cells stimulated with lipopolysaccharide (LPS), irrespective of diets (p < 0.05). The 19% fat diet resulted in increased plasma soluble intercellular adhesion molecule (sICAM)-1 after exercise. Leukotriene (LT) B4 concentration released by neutrophils stimulated with LPS was higher in the 50% fat diet, compared to the lower fat diets, and the sICAM-1 production of neutrophils stimulated with LPS was significantly increased after exercise only with 30% fat diet. CONCLUSION: While a short, intense bout of exercise increased pro-inflammatory mediators of the immune system, decreasing fat intake to 19% on a caloric deficient diet caused a greater increase in plasma TNF-alpha, sVCAM-1 and sICAM-1 concentration than the 30% and 50% fat diets in male and female subjects. Increasing fat calories to 50% with caloric balance did not exacerbate pro-inflammatory mediators compared to a 30% fat diet.

Side-chain anchoring strategy for solid-phase synthesis of peptide acids with C-terminal cysteine.

Many naturally occurring peptide acids, e.g., somatostatins, conotoxins, and defensins, contain a cysteine residue at the C-terminus. Furthermore, installation of C-terminal cysteine onto epitopic peptide sequences as a preliminary to conjugating such structures to carrier proteins is a valuable tactic for antibody preparation. Anchoring of N(alpha)-Fmoc, S-protected C-terminal cysteine as an ester onto the support for solid-phase peptide synthesis is known to sometimes occur in low yields, has attendant risks of racemization, and may also result in conversion to a C-terminal 3-(1-piperidinyl)alanine residue as the peptide chain grows by Fmoc chemistry. These problems are documented for several current strategies, but can be circumvented by the title anchoring strategy, which features the following: (a). conversion of the eventual C-terminal cysteine residue, with Fmoc for N(alpha)-amino protection and tert-butyl for C(alpha)-carboxyl protection, to a corresponding S-xanthenyl ((2)XAL(4)) preformed handle derivative; and (b). attachment of the resultant preformed handle to amino-containing supports. This approach uses key intermediates that are similar to previously reported Fmoc-XAL handles, and builds on earlier experience with Xan and related protection for cysteine. Implementation of this strategy is documented here with syntheses of three small model peptides, as well as the tetradecapeptide somatostatin. Anchoring occurs without racemization, and the absence of 3-(1-piperidinyl)alanine formation is inferred by retention of chains on the support throughout the cycles of Fmoc chemistry. Fully deprotected peptides, including free sulfhydryl peptides, are released from the support in excellent yield by using cocktails containing a high concentration (i.e., 80-90%) of TFA plus appropriate thiols or silanes as scavengers. High-yield release of partially protected peptides is achieved by treatment with cocktails containing a low concentration (i.e., 1-5%) of TFA. In peptides with two cysteine residues, the corresponding intramolecular disulfide-bridged peptide is obtained by either (a). oxidation, in solution, of the dithiol product released by acid; (b). simultaneous acidolytic cleavage and disulfide formation, achieved by addition of the mild oxidant DMSO to the cleavage cocktail; or (c). concomitant cleavage/cooxidation (involving a downstream S-Xan protected cysteine), using reagents such as iodine or thallium tris(trifluoroacetate) in acetic acid.

Effect of dietary intake on immune function in athletes.

Athletes are exposed to acute and chronic stress that may lead to suppression of the immune system and increased oxidative species generation. In addition, the tendency to consume fewer calories than expended and to avoid fats may further compromise the immune system and antioxidant mechanisms. The exercise stress is proportional to the intensity and duration of the exercise, relative to the maximal capacity of the athlete. Muscle glycogen depletion compromises exercise performance and it also increases the stress. Glycogen stores can be protected by increased fat oxidation (glycogen sparing). The diets of athletes should be balanced so that total caloric intake equals expenditure, and so that the carbohydrates and fats utilised in exercise are replenished. Many athletes do not meet these criteria and have compromised glycogen or fat stores, have deficits in essential fats, and do not take in sufficient micronutrients to support exercise performance, immune competence and antioxidant defence. Either overtraining or under nutrition may lead to an increased risk of infections. Exercise stress leads to a proportional increase in stress hormone levels and concomitant changes in several aspects of immunity, including the following: high cortisol; neutrophilia; lymphopenia; decreases in granulocyte oxidative burst, nasal mucociliary clearance, natural killer cell activity, lymphocyte proliferation, the delayed-type sensitivity response, the production of cytokines in response to mitogens, and nasal and salivary immunoglobulin A levels; blunted major histocompatibility complex II expression in macrophages; and increases in blood granulocyte and monocyte phagocytosis, and pro- and anti-inflammatory cytokines. In addition to providing fuel for exercise, glycolysis, glutaminlysis, fat oxidation and protein degradation participate in metabolism and synthesis of the immune components. Compromising, or overusing, any of these components may lead to immunosuppression. In some cases, supplementation with micronutrients may facilitate the immune system and compensate for deficits in essential nutrients. In summary, athletes should eat adequate calories and nutrients to balance expenditure of all nutrients. Dietary insufficiencies should be compensated for by supplementation with nutrients, with care not to over compensate. By following these rules, and regulating training to avoid overtraining, the immune system can be maintained to minimise the risk of upper respiratory tract infections.