Deep learning virtual indenter maps nanoscale hardness rapidly and non-destructively, revealing mechanism and enhancing bioinspired design

Over evolution, organisms develop complex material structures fit to their environments. Based on these time-tested designs, hu-man-engineered bioinspired structures offer exciting possible ma-terials configurations. However, navigating diverse structure spaces for attaining desired properties remains non-trivial. We focus on the hardest biological tissue in humans, tooth enamel, to examine the structure-property relationship. While typical hardness measure-ments are time consuming and destructive, we propose that artifi-cial intelligence models can predict properties directly and enable high-throughput, non-destructive characterization. We train a deep image regression neural network as a surrogate model and visualize with gradient ascent and saliency maps to identify struc-tural features contributing most to hardness. This model demon-strates improved spatial resolution and sensitivity compared with experimental hardness maps. Using this rapid hardness testing model, a generative adversarial model, and a genetic algorithm that operates in latent space, allows for guided materials design, yielding proposed designs for bioinspired structures with precisely controlled hardness.

Automated summary

Bioinspired structures created by human engineering offer exciting possibilities for material configurations, but attaining desired properties is still challenging. This study examines the structure-property relationship by focusing on tooth enamel, the hardest biological tissue in humans. The use of artificial intelligence models enables rapid and non-destructive characterization of properties, and a deep image regression neural network is trained as a surrogate model. This model improves spatial resolution and sensitivity compared to experimental hardness maps, allowing for guided materials design.