Explainable deep learning for diabetes diagnosis with DeepNetX2

Biomedical Signal Processing and Control, Год журнала: 2024, Номер 99, С. 106902 - 106902

Опубликована: Сен. 13, 2024

Язык: Английский

Towards Transparent Diabetes Prediction: Combining AutoML and Explainable AI for Improved Clinical Insights

Information, Год журнала: 2024, Номер 16(1), С. 7 - 7

Опубликована: Дек. 26, 2024

Diabetes is a global health challenge that requires early detection for effective management. This study integrates Automated Machine Learning (AutoML) with Explainable Artificial Intelligence (XAI) to improve diabetes risk prediction and enhance model interpretability healthcare professionals. Using the Pima Indian dataset, we developed an ensemble 85.01% accuracy leveraging AutoGluon’s AutoML framework. To address “black-box” nature of machine learning, applied XAI techniques, including SHapley Additive exPlanations (SHAP), Local Interpretable Model-Agnostic Explanations (LIME), Integrated Gradients (IG), Attention Mechanism (AM), Counterfactual Analysis (CA), providing both patient-specific insights into critical factors such as glucose BMI. These methods enable transparent actionable predictions, supporting clinical decision-making. An interactive Streamlit application was allow clinicians explore feature importance test hypothetical scenarios. Cross-validation confirmed model’s robust performance across diverse datasets. demonstrates integration pathway achieving accurate, interpretable models foster transparency trust while decisions.

Язык: Английский

iSee: A case-based reasoning platform for the design of explanation experiences

Knowledge-Based Systems, Год журнала: 2024, Номер 302, С. 112305 - 112305

Опубликована: Авг. 8, 2024

Explainable Artificial Intelligence (XAI) is an emerging field within (AI) that has provided many methods enable humans to understand and interpret the outcomes of AI systems. However, deciding on best explanation approach for a given problem currently challenging decision-making task. This paper presents iSee project, which aims address some XAI challenges by providing unifying platform where personalized experiences are generated using Case-Based Reasoning. An experience includes proposed solution particular explainability its corresponding evaluation, end user. The ultimate goal provide open catalog can be transferred other scenarios trustworthy required.

Язык: Английский