ABSTRACT
Artificial intelligence (AI) and digital twin technologies are reshaping precision oncology by enabling the creation of continuously evolving virtual representations of individual cancer patients capable of simulating disease progression, therapeutic response, toxicity, and clinical outcomes. Traditional cancer management is largely based on episodic clinical assessments and standardized treatment strategies that frequently fail to capture the dynamic biological evolution of tumour and the complexity of patient-specific responses. AI-driven cancer digital twins overcome these limitations by integrating multimodal biomedical information—including radiological imaging, digital pathology, genomics, transcriptomics, proteomics, metabolomics, epigenomics, spatial biology, laboratory biomarkers, liquid biopsy, wearable physiological monitoring, electronic health records, and longitudinal clinical data—into adaptive computational ecosystems. Foundation AI models, transformer architectures, graph neural networks, self-supervised learning, reinforcement learning, and generative artificial intelligence collectively enable these virtual patients to continuously learn from evolving biomedical evidence while supporting biomarker discovery, predictive therapeutics, immunotherapy optimization, adaptive treatment planning, disease monitoring, and intelligent clinical decision support. Emerging innovations including multimodal large language models, federated learning, retrieval-augmented generation, explainable artificial intelligence, and agentic AI further enhance the capabilities of digital twin ecosystems by enabling collaborative, privacy-preserving, and continuously adaptive computational oncology. Despite remarkable technological progress, important scientific, technical, ethical, and regulatory challenges remain regarding multimodal data harmonization, computational scalability, interpretability, interoperability, cybersecurity, clinical validation, and equitable implementation. This review provides a comprehensive overview of AI-driven cancer digital twins, emphasizing computational foundations, multimodal integration, clinical applications, implementation challenges, and future perspectives for establishing a new paradigm of personalized oncology.[1].
Keywords: Digital twins, Artificial intelligence, Precision oncology, Foundation models, Computational oncology, Personalized medicine, Multi-omics, Digital pathology, Clinical decision support, Predictive oncology.
