基于约束局部模型的全自动桡骨分割

Fully Automatic Segmentation of the Radius Using Constrained Local Model

骨龄是衡量少年儿童骨骼发育程度的重要指标.桡骨,作为骨龄评估中的特征骨块,其年龄对骨龄的预测具有重要意义.为预测桡骨骨龄首先需要准确分割桡骨,本文通过基于随机森林回归投票的约束局部模型算法,通过多形状模板的建立,实现桡骨的自动分割,为后续桡骨骨龄的预测提供可靠的依据.

Bone age is one of the important index measuring the degree of bone development in children and adolescents.Radius,as the characteristic bone in bone age assessment,its age is of great significance for the forecast of bone age.Predict radial bone age requires accurate segmentation of radius.In this paper,we use Constrained Local Model algorithm based on Random Forest regression voting to achieve segmentation,providing reliable basis for the prediction.