Label-free and rapid mechanics of single cells under high-density co-culture conditions by deep learning image recognition-assisted atomic force microscopy

• Developed a label-free AFM method for rapid single-cell mechanics under high-density co-culture using deep learning image recognition. • YOLOX neural network achieved highest detection precision (89.25%) among YOLO variants, enabling accurate cell type identification from bright-field images. • The method eliminates the need for fluorescent labeling, avoiding potential interference with cell behavior. • Enables efficient measurement of Young's modulus and adhesion forces of single cells in contact co-culture, advancing mechanobiology research.