While advanced imaging strategies have improved the diagnosis of bone-related pathologies, early signs of bone alterations remain difficult to detect. The Covid-19 pandemic has brought attention to the need for a better understanding of bone micro-scale toughening and weakening phenomena. This study used an artificial intelligence-based tool to automatically investigate and validate four clinical hypotheses by examining osteocyte lacunae on a large scale with synchrotron image-guided failure assessment. The findings indicate that trabecular bone features exhibit intrinsic variability related to external loading, micro-scale bone characteristics affect fracture initiation and propagation, osteoporosis signs can be detected at the micro-scale through changes in osteocyte lacunar features, and Covid-19 worsens micro-scale porosities in a statistically significant manner similar to the osteoporotic condition. Incorporating these findings with existing clinical and diagnostic tools could prevent micro-scale damages from progressing into critical fractures.
