The field of structural health monitoring (SHM) and materials science is witnessing significant developments, driven by the integration of advanced machine learning techniques and physics-informed approaches. Researchers are exploring new methods for in-service crack identification, damage detection, and materials design, leveraging techniques such as higher-order transmissibility, transfer learning, and graph neural networks. These innovations have the potential to enhance the accuracy and efficiency of SHM systems, as well as accelerate the discovery of new materials with desired properties. Noteworthy papers in this area include:

• A study proposing a new crack detection feature based on higher-order harmonics of the breathing crack, which can significantly reduce the computational burden of crack identification.
• A paper presenting a physics-informed transfer learning approach for SHM, which leverages knowledge of physics to select features that behave consistently across domains.
• A research work introducing a graph learning approach for metallic glass discovery, which uses Wikipedia embeddings and graph neural networks to explore hidden relationships among materials.
• A study developing a new semi-analytical model for dynamic analysis of free-free Timoshenko beams on elastic foundation under transverse transient ground deformation.
• A paper presenting a foundation model for material fracture prediction, which operates across simulators, materials, and loading conditions, and can be fine-tuned with minimal data on diverse downstream tasks.