From static to dynamic: An AI-integrated spatiotemporal model for early cancer detection via multi-omics liquid biopsy

Early diagnosis and timely intervention of cancer are crucial for reducing mortality rates. However, traditional diagnostic methods—such as imaging and tissue biopsy—have inherent limitations in sensitivity, invasiveness, and capturing tumor evolution dynamics, making them inadequate for early detection needs. Liquid biopsy, as a non-invasive technique, offers new avenues for deciphering early tumor mechanisms by analyzing biomarkers like circulating tumor DNA (ctDNA) and microRNA (miRNA). Advances in artificial intelligence (AI) and machine learning (ML) have further propelled breakthroughs in precision medicine, enabling the effective integration and interpretation of complex multi-omics data from liquid biopsies. This provides critical technological support for overcoming the limitations of traditional methods. This review proposes an AI/ML-driven integrated framework that synergistically analyzes the spatial genomic information of ctDNA with the temporal regulatory dynamics of miRNAs. This integration marks a fundamental shift from static, unidimensional detection paradigms to dynamic, multidimensional monitoring—transcending the limitations of single-analyte detection and simple multi-analyte complementarity, representing a paradigm innovation in early cancer detection. Specifically, we introduce an ML-driven “spatiotemporal complementary model” that fuses multidimensional biomarker data through advanced algorithms to generate a quantifiable “tumor evolution score” (TES) for dynamic cancer risk assessment. Furthermore, we delve into critical challenges in its clinical translation—including data standardization and model interpretability—and propose a clear implementation roadmap. By elucidating how interpretable AI frameworks can transform multi-omics liquid biopsies into precise early diagnostic and dynamic monitoring tools, this study aims to propel oncology from passive treatment models toward proactive, precision intervention strategies.