Deep learning-based multimodal cell elastography and analysis for quantification of cell mechanics

This study presents an intelligent multimodal cell elastography system that accurately measures the mechanical properties of both suspended and adherent cells. By integrating acoustic radiation force impulse (ARFI) stimulation with quantitative phase imaging (QPI), the system captures real-time cellular deformation with nanometer sensitivity. A deep learning-based analysis reconstructs deformation dynamics and extracts parameters such as stiffness and viscoelasticity. Applied to breast cancer cells, the system distinguished mechanical differences between highly invasive MDA-MB-231 and less invasive MCF-7 cells. This approach overcomes limitations of conventional acoustic tweezers and offers a robust, label-free tool for advanced cell mechanics and cancer research.