Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that affects millions of individuals worldwide, causing severe cognitive decline and memory impairment. The early and accurate diagnosis of AD is crucial for effective intervention and disease management. In recent years, deep learning techniques have shown promising results in medical image analysis, including AD diagnosis from neuroimaging data. However, the lack of interpretability in deep learning models hinders their adoption in clinical settings, where explainability is essential for gaining trust and acceptance from healthcare professionals. In this study, we propose an explainable AI (XAI)-based approach for the diagnosis of Alzheimer’s disease, leveraging the power of deep transfer learning and ensemble modeling. The proposed framework aims to enhance the interpretability of deep learning models by incorporating XAI techniques, allowing clinicians to understand the decision-making process and providing valuable insights into disease diagnosis. By leveraging popular pre-trained convolutional neural networks (CNNs) such as VGG16, VGG19, DenseNet169, and DenseNet201, we conducted extensive experiments to evaluate their individual performances on a comprehensive dataset. The proposed ensembles, Ensemble-1 (VGG16 and VGG19) and Ensemble-2 (DenseNet169 and DenseNet201), demonstrated superior accuracy, precision, recall, and F1 scores compared to individual models, reaching up to 95%. In order to enhance interpretability and transparency in Alzheimer’s diagnosis, we introduced a novel model achieving an impressive accuracy of 96%. This model incorporates explainable AI techniques, including saliency maps and grad-CAM (gradient-weighted class activation mapping). The integration of these techniques not only contributes to the model’s exceptional accuracy but also provides clinicians and researchers with visual insights into the neural regions influencing the diagnosis. Our findings showcase the potential of combining deep transfer learning with explainable AI in the realm of Alzheimer’s disease diagnosis, paving the way for more interpretable and clinically relevant AI models in healthcare.

Explainable artificial intelligence is beneficial in converting opaque machine learning models into transparent ones and outlining how each one makes decisions in the healthcare industry. To comprehend the variables that affect decision-making regarding diabetes prediction that can be accounted for by model-agnostic techniques. In this project, we investigate how to generate local and global explanations for a machine-learning model built on a logistic regression architecture. We trained on 253,680 survey responses from diabetes patients using the explainable AI techniques LIME and SHAP. LIME and SHAP were then used to explain the predictions produced by the logistic regression and Random forest-based model on the validation and test sets.With a discussion of future work, the comparative analysis and discussion of various experimental findings between LIME and SHAP are provided, along with their strengths and weaknesses in terms of interpretation. With a high accuracy of 86% on the test set, we used LR architecture with a spatial attention mechanism, demonstrating the possibility of merging machine learning and explainable AI to improve diabetes prediction, diagnosis, and treatment.We also focus on various applications, difficulties, and probable future directions of machine learning models for LIME and SHAP interpreters.

Objectives: The agricultural sector in many South Asian countries, including Bangladesh and India, plays a pivotal role in the economy, with a significant portion of the population dependent on it for livelihood. However, farmers often encounter challenges such as unpredictable weather conditions, soil variability, and natural disasters like floods and erosion, leading to substantial crop losses and financial strain. Despite government subsidies, many farmers struggle to sustain their livelihoods, resulting in a decline in interest in agriculture. Our focus lies on predicting the classification of various crops, including rice, jute, maize, and others, based on a combination of soil and weather features. Soil features, including Nitrogen, Phosphorus, Potassium, and pH levels, along with weather variables such as Temperature, Humidity, and Rainfall, are utilized as inputs for the predictive model. Methods: In this study, we address the critical issue of crop prediction by leveraging advanced machine-learning techniques and integrating genetic algorithms into the predictive model. Our proposed approach employs a hybrid methodology, where a Genetic Algorithm is utilized to optimize the hyperparameters of the model, enhancing its performance and robustness. Specifically, we employ a Random Forest classifier, a powerful ensemble learning technique, to classify the class labels associated with 22 different types of crops. Findings: The model’s accuracy is evaluated extensively, demonstrating a remarkable accuracy rate of 99.3%. Additionally, we utilized Local Interpretable Model-agnostic Explaination(LIME) and SHapley Additive exPlanations(SHAP) Explainable AI (XAI) methods to interpret and validate the model’s predictions. Novelty: The study presents a unique method for crop prediction that combines machine learning (ML) with genetic algorithms (GAs). The goal of this integration is to improve crop forecast models’ interpretability and accuracy. Due to the nature of local approximation LIME may yield contradictory answers. On the other hand, for sophisticated models and extensive datasets, SHAP can be computationally costly. By improving feature selection and model parameters, the integration of GAs with ML models overcomes these drawbacks and produces predictions that are more reliable and accurate. The high accuracy achieved by our system underscores its potential to mitigate crop losses and enhance agricultural productivity, thereby contributing to the sustainability and prosperity of the agricultural sector in any country.