Kälebo P, Strid K G
Department of Handicap Research, University of Gothenburg, Sweden.
Br J Radiol. 1989 Oct;62(742):883-9. doi: 10.1259/0007-1285-62-742-883.
A method for non-invasive assessment of experimental bone healing in animals is described. Optimized radiographs were analysed by digital processing using a computer-based image analysis system. Values of the total mass were expressed as the equivalent mass of a simultaneously exposed aluminium roentgen-density reference. In order to assess the potential of the method, measurements were initially performed on phantoms and bone specimens. The systematic error was negligible. An excellent correlation (r = 0.999) was found between the densitometric weight (mu) and the corresponding dry weight (m) of the analysed bone samples and a calibration factor mu/m = 1.38 was established. The total random error was low, the coefficient of variation (CV) amounting to less than 5% or, if duplicate analyses were performed, less than 3%. The accuracy was not impaired when radiography was carried out with addition of water to mimic overlying soft tissue in the model. The method was applied to the assessment of callus mass longitudinally in vivo in the healing of a bone defect in the rabbit radius, yielding reproducible values (CV less than 5%) and a quantitative analysis of the healing sequence.
描述了一种用于无创评估动物实验性骨愈合的方法。使用基于计算机的图像分析系统通过数字处理对优化后的X光片进行分析。总质量值表示为同时曝光的铝伦琴密度参考物的等效质量。为了评估该方法的潜力，最初对模型和骨标本进行了测量。系统误差可忽略不计。在分析的骨样本的密度测量重量（μ）和相应的干重（m）之间发现了极好的相关性（r = 0.999），并建立了校准因子μ/m = 1.38。总随机误差较低，变异系数（CV）小于5%，或者如果进行重复分析，则小于3%。当在模型中添加水以模拟覆盖的软组织进行X光摄影时，准确性不受影响。该方法被应用于纵向体内评估兔桡骨骨缺损愈合过程中的骨痂质量，得出可重复的值（CV小于5%）并对愈合过程进行了定量分析。
