Greedy Autoaugment for classification of mycobacterium tuberculosis image via generalized deep CNN using mixed pooling based on minimum square rough entropy.

Although tuberculosis (TB) is a disease whose cause, epidemiology and treatment are well known, some infected patients in many parts of the world are still not diagnosed by current methods, leading to further transmission in society. Creating an accurate image-based processing system for screening patients can help in the early diagnosis of this disease. We provided a dataset containing1078 confirmed negative and 469 positive Mycobacterium tuberculosis instances. An effective method using an improved and generalized convolutional neural network (CNN) was proposed for classifying TB bacteria in microscopic images. In the preprocessing phase, the insignificant parts of microscopic images are excluded with an efficient algorithm based on the square rough entropy (SRE) thresholding. Top 10 policies of data augmentation were selected with the proposed model based on the Greedy AutoAugment algorithm to resolve the overfitting problem. In order to improve the generalization of CNN, mixed pooling was used instead of baseline one. The results showed that employing generalized pooling, batch normalization, Dropout, and PReLU have improved the classification of Mycobacterium tuberculosis images. The output of classifiers such as Naïve Bayes-LBP, KNN-LBP, GBT-LBP, Naïve Bayes-HOG, KNN-HOG, SVM-HOG, GBT-HOG indicated that proposed CNN has the best results with an accuracy of 93.4%. The improvements of CNN based on the proposed model can yield promising results for diagnosing TB.

Plant dehydrins: shedding light on structure and expression patterns of dehydrin gene family in barley.

Dehydrins, an important group of late embryogenesis abundant proteins, accumulate in response to dehydration stresses and play protective roles under stress conditions. Herein, phylogenetic analysis of the dehydrin family was performed using the protein sequences of 108 dehydrins obtained from 14 plant species based on plant taxonomy and protein subclasses. Sub-cellular localization and phosphorylation sites of these proteins were also predicted. The protein features distinguishing these dehydrins categories were identified using various attribute weighting and decision tree analyses. The results revealed that the presence of the S motif preceding the K motif (YnSKn, SKn, and SnKS) was more evident and the YnSKn subclass was more frequent in monocots. In barley, as one of the most drought-tolerant crops, there are ten members of YnSKn out of 13 HvDhns. In promoter regions, six types of abiotic stress-responsive elements were identified. Regulatory elements in UTR sequences of HvDhns were infrequent while only four miRNA targets were found. Furthermore, physiological parameters and gene expression levels of HvDhns were studied in tolerant (HV1) and susceptible (HV2) cultivars, and in an Iranian tolerant wild barley genotype (Spontaneum; HS) subjected to gradual water stress and after recovery duration at the vegetative stage. The results showed the significant impact of dehydration on dry matter, relative leaf water, chlorophyll contents, and oxidative damages in HV2 compared with the other studied genotypes, suggesting a poor dehydration tolerance, and incapability of recovering after re-watering in HV2. Under severe drought stress, among the 13 HvDhns genes, 5 and 10 were exclusively induced in HV1 and HS, respectively. The gene and protein structures and the expression patterns of HvDhns as well as the physiological data consistently support the role of dehydrins in survival and recovery of barley plants from drought particularly in HS. Overall, this information would be helpful for functional characterization of the Dhn family in plants.