Synthetic data has emerged as a transformative tool in healthcare, particularly in areas such as medical imaging, electronic health records (EHRs), and clinical trial simulation, where data privacy, diversity, and accessibility are critical. This scoping review examines current approaches to synthetic data generation in healthcare, with a focus on AI model training, privacy preservation, and bias mitigation. A comprehensive search of PubMed, IEEE Xplore, and ACM Digital Library yielded 2,906 studies, of which 42 met the inclusion criteria. Key data generation techniques included generative adversarial networks (GANs), variational autoencoders (VAEs), diffusion models, Bayesian networks, federated learning, recurrent neural networks (RNNs), large language models (LLMs), agent-based models, graph-based generators, and SMOTE-based oversampling. Applications ranged from diagnostic model development to privacy-preserving data sharing and educational simulation. However, the field faces persistent challenges, including inconsistent validation practices, the absence of standard benchmarks, high computational demands, and ethical concerns related to consent and bias. This review underscores the need for standardized evaluation protocols, clearer regulatory guidance, and multidisciplinary collaboration to ensure the safe, equitable, and effective use of synthetic data in healthcare AI.

Objectives: Alzheimer’s disease (AD) remains one of the most prevalent neurodegenerative conditions among older adults, underscoring the urgent need for accurate and ethically grounded early detection methods. Artificial intelligence (AI) techniques, particularly machine learning and deep learning models, show promise in leveraging neuroimaging biomarkers to support early diagnosis. However, significant challenges persist regarding model explainability, accountability, and responsible implementation in real-world healthcare settings. This study presents a generalized Responsible AI (RAI) framework composed of four core components—explainability, fairness, predictive performance, and uncertainty quantification—to address these challenges. Method: Using the TADPOLE neuroimaging dataset, we implemented a Feedforward Neural Network (FNN) within a unified Responsible AI (RAI) framework integrating explainability, fairness, predictive performance, and uncertainty quantification. Although Random Forest achieved slightly higher predictive accuracy (95%), the FNN was selected as the primary model because it better supports end-to-end uncertainty estimation through Monte Carlo Dropout, enabling more reliable clinical decision support. Results: The proposed framework demonstrated strong predictive performance (92% accuracy), improved fairness reflected by an equalized odds difference of 0.124, and progressively lower predictive entropy across training iterations, indicating enhanced confidence in predictions. The framework further enabled model transparency through explainability analyses and supported the identification of low-confidence predictions for potential clinical review. Conclusions: Our findings highlight not only the feasibility of integrating RAI principles into AD prediction pipelines but also the persistent challenges of applying such frameworks to real-world clinical data. This work contributes practical insights toward operationalizing Responsible AI in healthcare contexts.