Classifying rice leaf diseases in agricultural technology helps to maintain crop health and to ensure a good yield. In this work, deep learning algorithms were, therefore, employed for the identification and classification of rice leaf diseases from images of crops in the field. The initial algorithmic phase involved image pre-processing of the crop images, using a bilateral filter to improve image quality. The effectiveness of this step was measured by using metrics like the Structural Similarity Index (SSIM) and the Peak Signal-to-Noise Ratio (PSNR). Following this, this work employed advanced neural network architectures for classification, including Cascading Autoencoder with Attention Residual U-Net (CAAR-U-Net), MobileNetV2, and Convolutional Neural Network (CNN). The proposed CNN model stood out, since it demonstrated exceptional performance in identifying rice leaf diseases, with test Accuracy of 98% and high Precision, Recall, and F1 scores. This result highlights that the proposed model is particularly well suited for rice leaf disease classification. The robustness of the proposed model was validated through k-fold cross-validation, confirming its generalizability and minimizing the risk of overfitting. This study not only focused on classifying rice leaf diseases but also has the potential to benefit farmers and the agricultural community greatly. This work highlights the advantages of custom CNN models for efficient and accurate rice leaf disease classification, paving the way for technology-driven advancements in farming practices.

Acute Lymphoblastic Leukemia (ALL), a cancer affecting the blood and bone marrow, requires precise classification for accurate diagnosis, personalized treatment plans, and improved predictive assessments to enhance patient survival and quality of life. This study presents LEU3, a novel classification model designed to improve the accuracy of leukemia detection from peripheral blood smear (PBS) images. LEU3 leverages an attention-based convolutional neural network (CNN) architecture, incorporating pooling layers, a global average pooling layer, and dense layers with dropout for regularization. The model is trained with an Adam optimizer comprising with four classes: Benign, early malignant pre-B, malignant pre-B, and malignant pro-B. Data augmentation techniques were employed to increase training set diversity. Additionally, Shapley Additive Explanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) are used to enhance interpretability and transparency in the model’s decision-making process. LEU3 achieved a test accuracy of 99% and a validation accuracy of 99% on 484 PBS images, demonstrating a 3% improvement over the baseline model. These results underline the potential of LEU3 in supporting medical professionals by reducing diagnostic workload and improving the accuracy of leukemia classification.

Guava (Psidium guajava) this is a tropical fruit and one of the common tropical fruits in Bangladesh. The economic and health value of this important crop is unmeasurable, but it quickly becomes infected with many diseases that can greatly reduce its yield and quality. Thus, the use of technology for automatic fruit and leaf disease detection is necessary in agriculture. The dataset provides us overall guava fruits & leaves image samples for detection of diseases in guava fruits and leaves. This dataset consists of images of healthy and diseased samples infected by fruit disease such as anthracnose, scab, styler end root and leaf disease such as canker, rust, anthracnose and dot. It consists of 3,432 real images obtained from different places in Bangladesh. It is also extended 20,344 augmented images ready to be used for machine learning purposes. This dataset serves as a fundamental building block for utilizing machine learning and computer vision techniques to develop automated detection systems of various diseases. It assists in the early detection of diseases affecting guava, and provides them with solutions to intervene there itself saving agricultural yield and nutritional losses while also promoting sustainable farming practices. This dataset will assist researchers for progressing guava detecting disease through the execution of computational models and application of better machine learning techniques.

Anemia is a critical medical condition in public health concern in tropical and subtropical areas, and understanding its hematological changes is crucial for improving diagnosis, treatment, and prognosis.It manifests through symptoms like weakness, fatigue, pale skin, and shortness of breath due to insufficient hemoglobin or red blood cells to carry adequate oxygen, with severe cases leading to complications such as chest pain. Common causes include blood loss, chronic diseases, and iron and vitamin deficiencies. This dataset captures various hematological parameters of patients suffering from anemia, including sex, age, Hemoglobin level (Hb), oxygen transportation (RBC), packed cell volume (PCV), mean corpuscular volume (MCV), Mean Corpuscular Hemoglobin (MCH), Mean corpuscular hemoglobin concentration (MCHC). The data is systematically collected from patients admitted to Aalok Healthcare Ltd., situated in Dhaka, Bangladesh., offering an opportunity to analyze hematological variations in patients with anemia. By providing a worldwide viewpoint for comparing hematological responses, this dataset aids in the development of prediction models for the severity of anemia and patient outcomes, improving clinical decision-making. The study examines how various treatment plans can affect blood characteristics, potentially leading to improved treatment strategies. For statistical analysis, the data is cleaning the noise (null and duplicate values), normalized, and encoded. The Chi-square test results indicate a p-value of $4.1929\times {10}^{-29}$, showing no significant association between gender and diagnostic outcomes. However, the Z-test and T-test results reveal a notable gender difference in hemoglobin levels, with p-values of $3.4789\times {10}^{-33}$ and $4.1586\times {10}^{-24}$, respectively, underscoring the relevance of gender in analyzing hemoglobin variations. These findings emphasize how gender influences hematological responses against Anemia. The dataset will greatly advance research on anemia, improve these critical medical terms in public health strategies, and enhance patient diagnosis and treatment methods, offering a distinct advantage.

This dataset on eggplant leaf diseases has been meticulously developed to provide a valuable resource for agricultural research and the advancement of automated disease detection systems. It comprises 4,089 high-resolution images of eggplant leaves, systematically categorized into six distinct classes: Healthy Leaf, Insect Pest Disease, Leaf Spot Disease, Mosaic Virus Disease, White Mold Disease, and Wilt Disease. The images were captured using smartphone cameras under controlled conditions with a consistent white background to ensure clarity and uniformity. To reflect real-world agricultural scenarios, data collection was conducted across multiple geographic locations and in varying lighting conditions. This approach enhances the dataset's diversity and applicability. The dataset underwent thorough manual labelling and preprocessing to ensure accuracy and consistency across all samples. Each image is clearly labelled according to its respective disease class, making the dataset readily usable for machine learning applications. The balanced representation of healthy and diseased leaves allows for comprehensive training and testing of classification models. Designed to support the development of machine learning models for the early detection and classification of eggplant diseases, this dataset holds significant reuse potential in various research domains. It is particularly suitable for applications in plant pathology, precision agriculture, and disease forecasting, where timely and accurate diagnosis is crucial. The dataset is freely available for academic and research purposes, making it a valuable resource for researchers and developers aiming to innovate in agricultural technology and crop management. With its robust design and practical focus, the dataset has the potential to drive advancements in sustainable farming practices and enhance agricultural productivity.

Leaf and fruit diseases are unavoidable due to the unpredictability of climate variability and environmental changes. This study presents dragon fruit and leaf diseases dataset. The dataset includes a total of 4,518 clear and detailed images. The dataset lists the following diseases of leaf and fruit along with the number of images for each: Bacterial Diseases (498), Fungal Infections (Anthracnose or Stem Canker) (84), Healthy Leaves (2,242), Sunburn Damage (271), Healthy Fruits (333), Insect Infected Fruits (53), Mealybugs and Scale Insects (Fruits) (173), Bacterial Wilt (Fruits) (84). All images were captured with two smartphones one is Google Pixel 7 Pro and another one is iPhone 14 Pro in the Dhaka Division. This dataset focuses on analysing the diseases of Dragon leaf and fruit to improving the agriculture sector.