Abstract It is widely accepted that bone mineral density affects outcomes for spinal arthrodesis surgeries. Traditional techniques such as dual-energy X-ray absorptiometry (DEXA) historically have been used to screen for osteopenia and osteoporosis. Hounsfield units (HUs) are emerging as a potentially superior metric to DEXA in osteoporosis identification and preoperative planning. HU measurements influence spinal instrumentation outcomes, with lower values correlating with postoperative complications such as pseudoarthrosis, interbody subsidence, and proximal junctional kyphosis. Advancements in artificial intelligence (AI) and machine learning (ML) expand the utility of CT HU to improve osteoporosis screening, predict fracture risk, and optimize surgical planning. This review aims to: (1) discuss the use of HU units in osteoporosis diagnosis, (2) summarize the correlation of HU and spinal surgery outcomes, and (3) present emerging AI and ML models that use segmentation technology and HU analysis to advance diagnosis and surgical planning.
Shanahan et al. (Tue,) studied this question.