摘要
组织损伤的无损检测是HIFU治疗过程中非常重要的辅助过程。提出了一种通过B超图像无损地检测组织损伤的方法。HIFU辐射后,组织损伤必然会引起B超图像的变化,只对HIFU焦域附近的数字减影图像进行处理和分析。利用新鲜离体猪肉组织和猪肝组织实验,利用主分量分析(PCA)和线性判别分析(LDA)相结合的方法,对减影图像进行特征提取,利用最近邻方法进行分类识别。对于离体猪肉组织中损伤的识别率达94.5%,对于离体猪肝组织中组织损伤的识别率达87.5%。PCA-LDA方法的识别率要优于平均灰度作为识别特征等利用单一特征的识别方法。
Noninvasive detection of lesion in tissue is an important aided way in the process of HIFU treatment. A method is proposed for noninvasive detection of lesion in tissue in HIFU. Lesion in tissue, caused by HIFU exposures, inevitably leads to change in B-mode ultrasound images. The difference-sub-image near focus of HIFU, where the lesion of tissue appears, was used. Features of difference-sub-images were extracted by combining the principal component analysis (PCA) and linear discriminant analysis (LDA), and lesion in tissue was recognized using the nearest neighbor algorithm. In vitro experiments on fresh pork and liver were conducted. The accuracy for recognizing lesion pork tissue was 94.5 %, and that of liver tissue was 87.5 %. The performance of the proposed method is better than that of other ones based on single feature extraction.
出处
《中国生物医学工程学报》
CAS
CSCD
北大核心
2008年第6期812-816,共5页
Chinese Journal of Biomedical Engineering
关键词
HIFU
组织损伤
主分量分析
线性判别分析
HIFU
lesion of tissue
principal component analysis
linear discriminant analysis