文章摘要

文章信息/Info

Title:: Ultrasonic Image Enhancement Algorithm for Grading Diagnosis of Fatty Liver

作者:: 王正发1; 栾强厚1; 黄鑫1; 彭国华3; 武剑辉1; 2△; 1.广东睿超电子科技有限公司，东莞 523808；2. 电子科技大学，成都 610054；3. 深圳迈瑞生物医疗电子股份有限公司，深圳 518057

Author(s):: WANG Zhengfa1; LUAN Qianghou1; HUANG Xin1; PENG Guohua3; WU Jianhui1; 2; 1.Guangdong Ruichao Electronic Technology Co. Ltd., Dongguan 523808，Guang Province，China； 2.University of Electronic and Technology of China, Chengdu 610054，China；3. Shenzhen Mindray Biological Medical Electronic Co., LTD., Shenzhen 518057，China

Keywords:: Fatty liver; Ultrasonic image; Resolution ratio; Enhancement algorithm; Piecewise linear transformation; Grading diagnosis

摘要:: 提出一种针对脂肪肝分级诊断的超声图像增强算法，提高超声图像对脂肪肝的分级诊断能力。通过分析正常肝和不同程度脂肪肝在超声图像上的表现特点，本研究提出了一种将灰度的分段线性变换方法和改进的局部对比度增强方法相结合的图像增强方法，该方法首先通过灰度分段线性变换处理来凸显图像的脂肪肝特征，然后利用改进的局部对比度增强处理来增强图像的细节，提高图像的整体质量。应用该方法分别对正常肝和轻度、中度以及重度脂肪肝超声图像进行增强处理并对比原图像，结果表明，增强后图像的脂肪肝特征更加凸显，图像细节得到了提高，不同程度脂肪肝图像之间的差异更明显，具有更好的脂肪肝诊断信息。本研究提出的方法可提高超声图像分级诊断脂肪肝的能力，具有一定的实际应用价值。

Abstract:: To propose a enhancement algorithm of ultrasonic image for improving the capability of grading diagnosis in fatty liver. The algorithm which combines the piecewise linear transformation of gray level was proposed and the local contrast ratio enhancement was improved, by analysing the characteristics of ultrasonic image in normal liver and fatty liver with different degree. The features of fatty liver in ultrasonic image were highlighted with the piecewise linear transformation of gray level. And the details of image were enhanced with the improved local contrast ratio enhancement processing while improving the quality of ultrasonic image. The ultrasonic images of normal liver, mild fatty liver, moderate fatty liver and severe fatty liver were processed with the enhancement algorithm. By comparing the original images, it could be found that the enhanced images had better features of fatty liver and better image details, and the differences in images of fatty liver with different degree were more obvious. It showed that the enhanced images had better information for diagnosing fatty liver. The enhancement algorithm proposed can improve the capability of grading diagnosis in fatty liver, has some references to practical application.