基于X光图像的桡骨等级评定方法研究

发布时间：2018-04-26 17:18

本文选题：中华05计分法 + 桡骨　；参考：《北京交通大学》2016年硕士论文

【摘要】：骨龄是衡量儿童青少年体格发育程度的重要数据指标,在临床医学、体育科学和司法等领域有着广泛的应用。由于手腕骨能够比较准确地反映全身骨骼生长发育状况,且易于拍摄,照射剂量小,因此国内外学者普遍认为手腕骨是确定骨龄的最理想部位。中华05计分法是我国目前识别骨龄所使用的手腕骨发育成熟度评价方法。等级评定是中华05计分法中最关键的步骤,它需要工作人员经过专业的培训,并熟练掌握各骨发育分期才能进行有效的人工判定,整个评定过程比较繁琐、工作量大、周期长。随着计算机技术的普及以及数字图像处理技术的发展,实现等级的自动评定成为了目前骨龄研究的热点。本文研究了一种基于中华05计分法的桡骨等级评定方法,针对手腕骨X光图像中桡骨的特点,对桡骨等级8-0至等级8-4的自动评定过程进行了研究。本文的主要内容如下：(1)提出了一种基于区域生长与最大类间方差法结合的桡骨分割算法。首先对手腕骨X光图像进行灰度变换、图像裁剪操作,然后对裁剪图像进行分块处理,采用最大类间方差法获得每块图像的自适应阈值,最后结合自适应阈值制定相似性准则,通过区域生长法实现手腕骨X光图像中桡骨的有效分割。(2)根据中华05计分法中桡骨等级8-0至8-4的评定标准以及手腕骨X光图像中桡骨的影像特点,首先选取了部分具有代表性的灰度特征和纹理特征,并提出了融合率新特征作为骨龄特征参数。然后分别设计k近邻分类器和随机森林分类器对桡骨等级进行分类实验,验证了选取的骨龄特征的分类性能。(3)基于ITK和VTK类库,结合MFC搭建了一个桡骨等级评定系统平台。通过对桡骨等级评定的过程进行研究,将整个系统分为图像读取,图像显示,图像交互,桡骨分割,等级自动评定五个功能模块。通过对各模块应用技术的分析,设计并实现了相应的功能。
[Abstract]:Bone age is an important data index to measure the physical development of children and adolescents. It is widely used in clinical medicine, sports science and justice. Because the wrist bone can accurately reflect the whole body bone growth and development, and is easy to photograph, the irradiation dose is small, so the wrist bone is generally considered to be the most ideal site to determine the bone age at home and abroad. Zhonghua 05 score method is a method to evaluate the maturity of wrist bone development used to identify bone age in China. Grade evaluation is the most critical step in the Chinese 05 score method. It requires the staff to undergo professional training and master the stages of bone development in order to make effective manual judgment. The whole evaluation process is more complicated, the workload is large, and the period is long. With the popularization of computer technology and the development of digital image processing technology, the realization of automatic grading has become the focus of bone age research. In this paper, a method of radial grade evaluation based on Zhonghua 05 scoring method is studied. According to the characteristics of radius in X-ray images of wrist bone, the automatic evaluation process of radius grade 8-0 to grade 8-4 is studied. The main contents of this paper are as follows: (1) A radial segmentation algorithm based on the combination of region growth and maximum inter-class variance is proposed. First of all, the opponent carpal X-ray image is transformed to gray level, the image is clipped, then the clipped image is processed in blocks, and the adaptive threshold of each image is obtained by using the maximum inter-class variance method. Finally, the similarity criterion is established by combining the adaptive threshold. The effective segmenting of radius in X-ray images of wrist bone was realized by region growth method.) according to the evaluation criteria of radius grade 8-0 to 8-4 in Zhonghua 05 score method and the characteristics of radius in X-ray images of wrist bone, Firstly, some representative grayscale features and texture features are selected, and a new feature of fusion rate is proposed as the feature parameter of bone age. Then we design k-nearest neighbor classifier and random forest classifier to classify radial grade, and verify the classification performance of the selected bone age feature. Based on ITK and VTK class library, a radial grading system platform is built with MFC. The whole system is divided into five functional modules: image reading, image display, image interaction, radius segmentation and grade automatic evaluation. By analyzing the application technology of each module, the corresponding functions are designed and realized.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：R816.8

【相似文献】