基于最大类间方差与形态学的淋巴结图像分割

Lymph Node Image Segmentation Algorithm Based on Maximal Variance Between-class and Morphology

淋巴结是人体内产生免疫应答的重要器官。淋巴结的病理变化是检测恶性肿瘤(肺癌、直肠癌、乳腺癌、肝癌、宫颈癌等)和判断肿瘤转移的一个重要依据。一种基于最大类间方差法与数学形态学的分割方法用于淋巴结的分割。最大类间方差法用于对原图进行二值化增强处理,而数学形态学方法用于修正二值图像的边界,通过腐蚀操作与膨胀操作解决二值化后出现的目标区域与多余组织相连的问题,以更好地提取有用的淋巴结组织。实验结果表明,上述算法对与周围组织有粘连但目标与背景的灰度级相差较大的淋巴结图像的分割效果较好。

Lymph nodes are an important organ of the human body immune response. The pathological changes of lymph node are an important basis of malignant tumor detection and judgment of metastasis of cancer （lung cancer, colorectal cancer, breast cancer, liver cancer, cervical cancer, etc. ） A segmentation algorithm based on maximal variance between-class and morphology was introduced to segment lymph node. Maximal variance between-class method was used to operate binary enhancement processing for the original image. Mathematical morphology was introduced to do boundary correction for binary image. Erosion and expansion operation was used to solve the problem of the target area connected with excess tissue after binarization. In the end, useful lymph node tissue was better extracted. Experimental results show that the proposed method can get better segmentation effect for lymph node images with surrounding tis- sue adhesions and larger gray level difference between target and background.