目的　通过对腰椎退变黄韧带的数字化重建，为影像诊断和临床手术规划，提供立体直观的形态学依据。   方法　收集腰腿痛患者薄层螺旋CT影像资料，在后处理工作站中观察退变黄韧带形态及分布，选取轴位像中黄韧带厚度大于4 mm和/或有钙化患者63例，男31例，女32例，年龄32~80岁。将原始数据导入Mimics16.0软件重建图像，观察其形态、分布。   结果　315个腰椎节段中205个节段出现黄韧带肥厚和/或骨化。黄韧带单纯肥厚占总出现节段79.51%，单纯骨化占6.34%，肥厚伴骨化占14.15%。205个节段全部成功建模，肥厚黄韧带建模形态有合页状、不规则片状和柱状；骨化黄韧带建模形态有短柱状、短锥状、不规则片状；肥厚伴骨化黄韧带重建后形态各异。  结论　数字化技术重建黄韧带能清楚显示其分布、形态、邻近结构，还可将所建退变黄韧带模型与腰椎骨性模型随意组合、切割，便于立体直观地显示病变细节。

Objective　Through the digital reconstruction of the degenerative of lumbar ligamentum flavum，a stereoscopical and intuitive morphologic basis for the degenerative lumbar ligamentum flavum for diagnosis and surgical planning was provided.　Methods　CT imaging data of lumbar pain patients were collected, and cross-sectional images of lumbar ligamentum flavum in post-processing workstation were observed, a total of 63 patients aged 32 to 80 years old whose lumbar ligamentum flavum either hypertrophied greater than 4 mm and / or ossified were selected, male 31 male  and 32 female. Then, CT scan raw data were imported into mimics 16.0 software for image reconstruction, its morphology and distribution were observed.　Results　In the 63 cases of 315 lumbar vertebrae, a total of 205 segments’ ligamentum flavum were hypertrophied and /or ossified. The ligamentum flavum hypertrophied alone was 79.51%, and ossification alone ossification was 6.34%. Hypertrophy with ossification was 14.15%. All 205 segments’ ligamentum flavum were successfully modelled and ligamentum flavum hypertrophy modeling morphology included hinge-like, irregular flake and irregular column, ossification of ligamentum flavum contained with short columnar shape, short cone-shape, irregular flake, ligamentum flavum hypertrophied with ossification had their different forms after reconstruction.　Conclusion　The computer digital technology reconstruction of lumbar degenerative ligamentum flavum could clearly demonstrate the distribution, morphology and adjacent structure of the relationship. Moreover, lumbar spinal models and computer constructed ligamentum flavum models could be randomly assembled and sectioned, which could comprehensively and intuitively display all details of the lesion.