Introduction: Diabetes is considered one of the leading healthcare concerns affecting millions worldwide. Taking appropriate action at the earliest stages of the disease depends on early diabetes prediction and identification. To support healthcare providers for better diagnosis and prognosis of diseases, machine learning has been explored in the healthcare industry in recent years.Methods: To predict diabetes, this research has conducted experiments on five boosting algorithms on the Pima diabetes dataset. The dataset was obtained from the University of California, Irvine (UCI) machine learning repository, which contains several important clinical features. Exploratory data analysis was used to identify the characteristics of the dataset. Moreover, upsampling, normalisation, feature selection, and hyperparameter tuning were employed for predictive analytics.Results: The results were analysed using various statistical/machine learning metrics and k-fold cross-validation techniques. Gradient boosting achieved the greatest accuracy rate of 92.85% among all the classifiers. Precision, recall, f1-score, and receiver operating characteristic (ROC) curves were used to further validate the model.Discussion: The suggested model outperformed the current studies in terms of prediction accuracy, demonstrating its applicability to other diseases with similar predicate indications.

The cultivation of most crops depends upon the regional weather conditions. So, the analysis of the agro-climatic conditions of a zone contributes significantly to deciding the right crop for the right land in the right season to obtain a better yield. Machine learning algorithms facilitate this process to a great extent for better results. In this paper, the authors proposed an ML-based crop selection model based on the weather conditions and soil parameters, collectively. Weather analysis is done using LSTM RNN and the process of crop selection is completed using Random Forest Classifier. This model gives better results for weather prediction in comparison to ANN. With LSTM RNN, the RMSE observed in Min. Temp. prediction is 5.023%, Max. Temp. Prediction is 7.28%, and Rainfall Prediction is 8.24%. In the second phase, the Random Forest Classifier showed 97.235% accuracy for crop selection, 96.437% accuracy in predicting resource dependency, and 97.647 accuracies in giving the appropriate sowing time for the crop. The model construction time taken with a random forest classifier using mentioned data size is 5.34 s. The authors also suggested the future research direction to further improve this work.

The neurodevelopmental Autism Spectrum Disorder (ASD) causes problems in social communication. Earlier diagnosis of ASD from brain image is necessary for reducing the effect of disorder. In this paper, deep Convolutional Neural Network (CNN) with Dwarf Mongoose optimized Residual Network (DM-ResNet) is proposed for the classification of autism disorder from Magnetic Resonance Imaging (MRI) brain images. Initially, the input brain images are preprocessed to remove the non-brain tissues. The preprocessed images are segmented with hybrid Fuzzy C Means (FCM) and Gaussian Mixture Model (GMM) which partition the image into sub groups to make it easier for classification by reducing the complexity. FCM-GMM segments the volume into predefined cortical and sub cortical regions. After segmentation, the features are extracted with Visual Geometry Group (VGG)-16 networks which comprised of several tiny kernels with filters for enhancing the depth of network and permit to extract complicated and discriminative features. Region of Interest (ROI) based functional connectivity feature is extracted with VGG-16 and these features are classified with DM optimized ResNet. The hyper parameters are optimized with DM optimization algorithm which improves the accuracy of classifier. By using the proposed approach, the accuracy of autism detection is improved to 99.83%.

It has been challenging to simultaneously improve photosynthesis and stress tolerance in plants. Crassulacean acid metabolism (CAM) is a CO₂-concentrating mechanism that facilitates plant adaptation to water-limited environments. We hypothesized that the ectopic expression of a CAM-specific phosphoenolpyruvate carboxylase (PEPC), an enzyme that catalyzes primary CO₂ fixation in CAM plants, would enhance both photosynthesis and abiotic stress tolerance. To test this hypothesis, we engineered a CAM-specific PEPC gene (named AaPEPC1) from Agave americana into tobacco. In comparison with wild-type and empty vector controls, transgenic tobacco plants constitutively expressing AaPEPC1 showed a higher photosynthetic rate and biomass production under normal conditions, along with significant carbon metabolism changes in malate accumulation, the carbon isotope ratio δ¹³C, and the expression of multiple orthologs of CAM-related genes. Furthermore, AaPEPC1 overexpression enhanced proline biosynthesis, and improved salt and drought tolerance in the transgenic plants. Under salt and drought stress conditions, the dry weight of transgenic tobacco plants overexpressing AaPEPC1 was increased by up to 81.8% and 37.2%, respectively, in comparison with wild-type plants. Our findings open a new door to the simultaneous improvement of photosynthesis and stress tolerance in plants.

Deepfake technology uses auto-encoders and generative adversarial networks to replace or artificially construct fine-tuned faces, emotions, and sounds. Although there have been significant advancements in the identification of particular fake images, a reliable counterfeit face detector is still lacking, making it difficult to identify fake photos in situations with further compression, blurring, scaling, etc. Deep learning models resolve the research gap to correctly recognize phony images, whose objectionable content might encourage fraudulent activity and cause major problems. To reduce the gap and enlarge the fields of view of the network, we propose a dual input convolutional neural network (DICNN) model with ten-fold cross validation with an average training accuracy of 99.36 ± 0.62, a test accuracy of 99.08 ± 0.64, and a validation accuracy of 99.30 ± 0.94. Additionally, we used ’SHapley Additive exPlanations (SHAP) ’ as explainable AI (XAI) Shapely values to explain the results and interoperability visually by imposing the model into SHAP. The proposed model holds significant importance for being accepted by forensics and security experts because of its distinctive features and considerably higher accuracy than state-of-the-art methods.

Technology offers a lot of potential that is being used to improve the integrity and efficiency of infrastructures. Crack is one of the major concerns that can affect the integrity or usability of any structure. Oftentimes, the use of manual inspection methods leads to delays which can worsen the situation. Automated crack detection has become very necessary for efficient management and inspection of critical infrastructures. Previous research in crack detection employed classification and localization-based models using Deep Convolutional Neural Networks (DCNNs). This study suggests and compares the effectiveness of transfer learned DCNNs for crack detection as a classification model and as a feature extractor to overcome this restriction. The main objective of this paper is to present various methods of crack detection on surfaces and compare their performance over 3 different datasets. Experiments conducted in this work are threefold: initially, the effectiveness of 12 transfer learned DCNN models for crack detection is analyzed on three publicly available datasets: SDNET, CCIC and BSD. With an accuracy of 53.40%, ResNet101 outperformed other models on the SDNET dataset. EfficientNetB0 was the most accurate (98.8%) model on the BSD dataset, and ResNet50 performed better with an accuracy of 99.8% on the CCIC dataset. Secondly, two image enhancement methods are employed to enhance the images and are transferred learned on the 12 DCNNs in pursuance of improving the performance of the SDNET dataset. The results from the experiments show that the enhanced images improved the accuracy of transfer-learned crack detection models significantly. Furthermore, deep features extracted from the last fully connected layer of the DCNNs are used to train the Support Vector Machine (SVM). The integration of deep features with SVM enhanced the detection accuracy across all the DCNN-dataset combinations, according to analysis in terms of accuracy, precision, recall, and F1-score.