Artificial intelligence application in multiscale biomechanics

• Outlined the fundamental principles and typical models of AI. • Introduced typical applications of AI in multiscale biomechanics. • Discussed the existing issues and challenges faced by current AI technologies in biomechanics. The rapid development and widespread application of artificial intelligence (AI) technology have significantly improved understanding across various fields, including biomechanics. To deepen the understanding of AI applications in this field and explore future developments, this review focuses on the progress of AI in multiscale biomechanics. We first outline the progress history, fundamental principles, and typical models of AI. Next, we introduce the main applications of the two typical AI paradigms—data-driven and knowledge-driven—in the context of multiscale biomechanical studies. The first paradigm focuses primarily on predicting protein structure, interactions, and conformational dynamics at the molecular level, as well as on subcellular structure recognition, cell mechanics prediction and cell trajectory tracking at the cellular level. The second paradigm concentrates on biological fluid and solid mechanics at the tissue level. Finally, the existing issues and challenges faced by current AI technologies in biomechanics are discussed, and potential future issues are proposed from the perspective of informative integration.