Walker Corey T, Babadjouni Robin, Gibbs Wende, Lord Elizabeth, Gausper Adeesya, Osorio Joseph, Molina Camilo, Jones Kristen, van Hooff Miranda, Theologis Alexander, Yagi Mitsuru, Blakemore Laurel, Shah Suken, Hu Serena, de Kleuver Marinus, Pizones Javier, Kelly Michael, Pellise Ferran, Ames Christopher, Eastlack Robert
Department of Neurosurgery, Cedars-Sinai Medical Center, Los Angeles, CA.
Department of Neuroradiology, Barrow Neurological Institute, Phoenix, AZ.
Spine (Phila Pa 1976). 2025 Sep 15;50(18):1278-1289. doi: 10.1097/BRS.0000000000005426. Epub 2025 Jun 13.
Narrative review.
To explore the intersection of osteoporosis, sarcopenia, radiomics, and machine learning in spine surgery, with a focus on clinical applications and opportunities for advancing assessment and predictive modeling methods.
Osteoporosis and sarcopenia are significant contributors to negative outcomes in the aging adult spine. Current methodologies for evaluating these disease states remain limited, with significant variability and poor standardization. Advances in computational medicine provide a novel opportunity to improve quantitative assessment of osteosarcopenia, as demonstrated in other areas of medicine. Using radiomic approaches for predictive outcome modeling in spine surgery remains largely untapped.
A comprehensive literature search was performed. Articles were identified using the search terms "osteoporosis," "sarcopenia," "osteosarcopenia," "radiomics," "spine surgery," and "machine learning." Relevant studies were selected based on their focus on the intersection of these topics, emphasizing clinical, imaging, and computational methodologies in spine surgery.
This review highlights the existing conventional and research methods of assessing both osteoporosis and sarcopenia, particularly regarding their clinical application in spine surgery. Areas of research within the radiomic space for both conditions are also discussed to describe opportunities for growth of future research and areas of focus needed to advance the field of spine surgery alongside the rapid growth of artificial intelligence.
Understanding the relationship between osteoporosis, sarcopenia, and frailty is essential to improving outcomes in spine surgery. Advanced imaging and machine learning approaches offer the potential for more precise assessments and tailored interventions. The Scoliosis Research Society Adult Spinal Deformity Task Force on Senescence has identified this as an area of maximal importance for strategic growth and development of the field.
叙述性综述。
探讨骨质疏松症、肌肉减少症、放射组学和机器学习在脊柱手术中的交叉点，重点关注临床应用以及推进评估和预测建模方法的机会。
骨质疏松症和肌肉减少症是导致老年成人脊柱不良结局的重要因素。目前评估这些疾病状态的方法仍然有限，存在显著的变异性且标准化程度差。计算医学的进展为改善骨质疏松症的定量评估提供了新的机会，这在医学的其他领域已得到证明。在脊柱手术中使用放射组学方法进行预测结果建模在很大程度上尚未得到充分利用。
进行了全面的文献检索。使用搜索词“骨质疏松症”“肌肉减少症”“骨质疏松性肌肉减少症”“放射组学”“脊柱手术”和“机器学习”来识别文章。根据对这些主题交叉点的关注，选择相关研究，重点是脊柱手术中的临床、影像和计算方法。
本综述强调了评估骨质疏松症和肌肉减少症的现有传统方法和研究方法，特别是它们在脊柱手术中的临床应用。还讨论了这两种情况在放射组学领域的研究领域，以描述未来研究的增长机会以及随着人工智能的快速发展推进脊柱手术领域所需关注的领域。
了解骨质疏松症、肌肉减少症和衰弱之间的关系对于改善脊柱手术的结局至关重要。先进的影像和机器学习方法为更精确的评估和个性化干预提供了潜力。脊柱侧凸研究协会衰老成人脊柱畸形特别工作组已将此确定为该领域战略增长和发展的一个极其重要的领域。
