纹理分析在甲状腺结节影像学中的研究进展

Research Progress of Texture Analysis in Thyroid Nodule Imaging

甲状腺结节是临床上常见疾病且多数为良性结节。影像检查的主要目的是筛查出恶性结节以及对恶性结节进行评估。纹理分析是近年来新出现的图像后处理技术,使用了各种数学方法来描述图像的不均匀性,通过定量参数对图像进行评估。纹理分析对于鉴别良恶性肿瘤、预测肿瘤分期及疗效评估等都具有一定的价值。本文就甲状腺结节的纹理分析研究进展予以综述。

Thyroid nodules are common diseases in clinic and are mostly benign nodules. The main purpose of imaging examination is to screen out and evaluate the malignant nodules. Texture analysis is a newly emerged image post-processing technology in recent years. Various mathematical methods are used to describe the nonuniformity of the image and the image is evaluated through quantitative parameters. Texture analysis has certain value in distinguishing benign and malignant tumors, predicting tumor stage and evaluating curative effect. The author summarizes the research progress in texture analysis of thyroid nodules.