AI Guided Measurement of Live Cells Using AFM

Atomic force microscopy (AFM), a member of the ‘scanning probe microscopy’ family, is an excellent platform for high-resolution imaging and mechanical characterization of a wide range of samples, including live cells, proteins, and other biomolecules. AFM is also beneficial for measuring interaction forces and binding kinetics for protein-protein or receptor-ligand interactions on live cells at a single-molecule level. However, high-resolution imaging and force measurements performed with AFM and data analytics are time-consuming and require specific skill sets and constant human supervision. It also involves problems such as cantilever tip breakage after prolonged functionalization and damage to the live cell samples due to lack of optimization of the loading forces, making this a low throughput method. Although remarkable progress in the area of AI and machine learning (ML) over the past few years has left its mark on bio-imaging as well, the potential of AI-AFM strategies in a live cell characterization has been mostly unexplored. In this paper, we developed an ML framework to perform automatic sample selection for AFM navigation during AFM biomechanical mapping. We established ML-based closed-loop scanner trajectory and force tracking algorithms for precise AFM positioning during sample navigation and biomechanical mapping at high speed. Our innovation will directly address state-of-the-art AFM operation via AI-driven intelligent automation, including intelligent navigation and image data analysis.