@Article{cmes.2026.088032,
AUTHOR = {Abdelzahir Abdelmaboud, Ibrahim Hameed, Salmah Fattah, Ashraf Osman Ibrahim, Salil Bharany, Ateeq Ur Rehman, SeongKi Kim},
TITLE = {Explainable, Context-Aware Artificial Intelligence and Digital Twin in Healthcare: A Systematic Review and Unified Taxonomy},
JOURNAL = {Computer Modeling in Engineering \& Sciences},
VOLUME = {148},
YEAR = {2026},
NUMBER = {3},
PAGES = {0--0},
URL = {http://www.techscience.com/CMES/v148n3/69012},
ISSN = {1526-1506},
ABSTRACT = {This paper provides a thorough, systematic review of explainable, context-sensitive Artificial Intelligence (AI) and Digital Twin (DT) technologies in healthcare, emphasizing their potential to transform the medical system into an intelligent, patient-centred system of care. The PRISMA framework helped identify 1842 records from the largest scientific databases, of which 97 high-quality studies were selected for a comprehensive analysis. The review analyses in detail the development of AI from traditional machine learning to modern deep learning and generative models with a focus on their use in disease diagnosis and predictive analytics, personalized medicine, and healthcare management. One of the priorities is the incorporation of Explainable AI (XAI), which tackles the significant issue of the black-box character of AI by making it more transparent, interpretable, and trusted by the clinical decision-making process. In addition, the paper examines how DT technology can be used to produce dynamic virtual models of patients and healthcare systems that can be monitored in real time, simulated, and analyzed to predict outcomes. An alternative, multi-layered taxonomy is proposed to bring together AI, XAI, DTs, and other emerging technologies, such as Internet of Things (IoT) and federated learning, within a unified, context-specific healthcare framework. The results show that although these technologies greatly enhance diagnostic accuracy, operational efficiency, and personalized care, difficulties such as data privacy concerns, model bias, lack of standardization, and high computational complexity prevent their adoption. The review finds significant research gaps, especially the lack of unified architectures, real-time explainability, and large-scale clinical validation. The work provides a systematic basis for future studies by describing the major challenges, innovation opportunities, and strategic paths to create integrated, understandable, and scalable intelligent healthcare ecosystems.},
DOI = {10.32604/cmes.2026.088032}
}