The accurate identification of diseases on apple production is an important issue due to the worldwide importance ofapple production in contemporary agriculture. Identifying diseases correctly can be challenging and affects food safety and economic losssignificantly. To alleviate this, deep learning approaches, and particularly Convolutional Neural Networks (CNN), have been able to providenew and reasonable options in the agricultural field. In this study, there is a hybrid model proposed, called DenseNet-ResNet-Hybrid,which brings together architectures from DenseNet and ResNet, to provide an improvement in the extraction of features together. It hasbeen designed to fuse the inherent capabilities of DenseNet and ResNet, capturing both detail features and deeper level features inapple images, to enhance the ability to separate diseases that are overlapped with the producer’s natural environment (e.g. overlappingleaves/fruits). We finally show two complete comparative experiments against two popular models (like VGG16, ResNet50, Inception-v3)under the exact same conditions to demonstrate the strength of their ability to accurately classify apple leaf diseases with consistency.We use a broader select of image types to demonstrate our work, and ultimately suggest our proposed hybrid model demonstratescompetitive performance in accurate classification on apple images on the whole.

This research examined the feasibility of utilizing ML algorithms to improve the initial detection and classification ofanemia and other blood disorders. The following study employed several traditional machine learning models: additional ML and AImethods were subsequently evaluated including - LightGB, CatBoost, Decision Tree, Gradient Boosting, Random Forest and XGBoost toblood-based features (RBC, WBC, HGB, and PLT). The results demonstrated that LightGB had the highest accuracy of 98.38%, thenfollowed by CatBoost at 98.37%. The Decision Tree and Gradient Boosting models respectively demonstrated an accuracy of 98.05%.The accuracy of Random Forest and XGBoost was 97.72%. These results show the possibility of ML techniques being able to uncoverhigher-level complex patterns in medical data to improve accuracy, particularly for anemia. The study presented new evidence andbaseline models to promote ML to expedite clinical decision making to provide timely intervention and develop personalized health care.The study provided evidence and potential usages for ML models to enable better clinical decision and action. The findings of this studyexplained that in the future using advanced technologies or deep learning, or addressing concerns relating to explainable AI methods, thecapabilities in clinical use should be optimized and expanded.

Brain tumor occurs when cells formed as a result of self-renewal of cells in the human body growmore than normal and become a mass. Brain tumor constitutes one of the factors that endanger human life.By early diagnosis with the right methods and techniques, lives can be saved by preventing brain tumors thatendanger human life. In today’s technology, Magnetic Resonance imaging (MRI) is used to detect brain tumors.Early diagnosis plays an important role in brain tumor. In this study, Convolution neural network (CNN) is usedfor brain tumor detection and classification with deep learning, a sub-branch of machine learning. When theCNN model was compared with other deep learning models for brain tumor prediction, it was found that theCNN model had a higher accuracy rate than other models, with 98.24%.

Early diagnosis of lung cancer is critical for improving patient prognosis. While Computer-Aided Diagnosis (CAD) systemsleveraging deep learning have shown promise, the selection of an optimal model architecture remains a key challenge. This study presentsa comparative analysis of three prominent Convolutional Neural Network (CNN) architectures InceptionV4, VGG-13, and ResNet-50 todetermine their effectiveness in classifying lung cancer into benign, malignant, and normal categories from Computed Tomography (CT)images. Utilizing the publicly available IQ-OTH/NCCD dataset, a transfer learning approach was employed, where models pre-trainedon ImageNet were fine-tuned for the specific classification task. To mitigate overfitting and enhance model generalization, a suite ofdata augmentation techniques was applied during training. It achieved an accuracy of 98.80%, with a precision of 98.97%, a recall of96.30%, and an F1-score of 97.52%. Notably, the confusion matrix analysis revealed that InceptionV4 perfectly identified all malignantand normal cases in the test set, highlighting its clinical reliability. The study also evaluated the trade-off between diagnostic performanceand computational efficiency, where InceptionV4 provided an optimal balance compared to the computationally intensive VGG-13 and theless accurate, albeit more efficient, ResNet-50. Our findings suggest that the architectural design of InceptionV4, with its multi-scalefeature extraction, is exceptionally well-suited for the complexities of lung cancer diagnosis. This model stands out as a robust and highlyaccurate candidate for integration into clinical CAD systems, offering significant potential to assist radiologists and improve early detectionoutcomes.