改进型脉冲耦合神经网络检测乳腺肿瘤超声图像感兴趣区域

汪源源; 焦静

doi:10.3788/OPE.20111906.1398

您当前的位置：

首页 >

文章列表页 >

改进型脉冲耦合神经网络检测乳腺肿瘤超声图像感兴趣区域

信息科学 | 更新时间：2020-08-12

- 改进型脉冲耦合神经网络检测乳腺肿瘤超声图像感兴趣区域
- Detection of regions of interest from breast tumor ultrasound images using improved PCNN
- 光学精密工程 2011年19卷第6期页码：1398-1405
- 作者机构：
  
  复旦大学电子工程系上海,200433
- 作者简介：
- 基金信息：
  
  国家自然科学基金资助项目(No.10974035)();上海市优秀学科带头人计划资助项目(No.10XD1400600)()
- DOI：10.3788/OPE.20111906.1398
  中图分类号： R445.1;TP183
- 收稿日期：2010-10-18，
  
  修回日期：2011-02-15，
  
  网络出版日期：2011-06-25，
  
  纸质出版日期：2011-06-25
- 稿件说明：
移动端阅览
汪源源, 焦静. 改进型脉冲耦合神经网络检测乳腺肿瘤超声图像感兴趣区域[J]. 光学精密工程, 2011,19(6): 1398-1405

WANG Yuan-yuan, JIAO Jing. Detection of regions of interest from breast tumor ultrasound images using improved PCNN[J]. Editorial Office of Optics and Precision Engineering, 2011,19(6): 1398-1405
汪源源, 焦静. 改进型脉冲耦合神经网络检测乳腺肿瘤超声图像感兴趣区域[J]. 光学精密工程, 2011,19(6): 1398-1405 DOI： 10.3788/OPE.20111906.1398.

WANG Yuan-yuan, JIAO Jing. Detection of regions of interest from breast tumor ultrasound images using improved PCNN[J]. Editorial Office of Optics and Precision Engineering, 2011,19(6): 1398-1405 DOI： 10.3788/OPE.20111906.1398.

摘要

为了解决超声图像斑点噪声、伪影、低图像对比度和图像亮度不均匀等问题

提出了一种改进的简化脉冲耦合神经网络(SPCNN)结合模糊互信息量的方法来自动检测乳腺肿瘤超声图像的感兴趣区域(ROI)。首先

对超声图像进行模糊增强预处理;然后

通过改进SPCNN对超声图像进行点火

以最大模糊互信息量作为最优判决准则

获得相应的分类结果;最后

对分类后的二值图像进行形态学等处理

从而得到乳腺超声图像的ROI。对包含118幅乳腺肿瘤超声图像的数据库进行处理

结果表明

该方法自动识别ROI准确率达到87.3%

处理每一幅图像的平均时间为4.68 s。本算法能有效快速地检测乳腺肿瘤超声图像的ROI

有望用于基于超声图像的乳腺肿瘤CAD中。

Abstract

To solve the problems of the speckle noise

pseudo image

low contrast and luminous inhomogeneity in an ultrasound image

a method based on the improved Simplified Pulse Coupled Neural Network (SPCNN) combined with the fuzzy mutual information model was proposed to detect the Region of Interest(ROI) of the breast tumor ultrasound image. The ultrasound image was firstly mapped to the fuzzy sets to enhance the contrast

then the SPCNN model was used to pulse the ultrasound image

and the fuzzy mutual information was used as the optimization criterion to obtain the relative classification results. The ROI of the breast tumor ultrasound image was finally obtained by applying the morphologic processing on the corresponding classified results. The proposed segmentation method was performed on 118 breast tumor ultrasound images

and the obtained results show that the ROI accuracy is 87.3% and average processing time per image is 4.68 s. In conclusion

the proposed meth-od can be used to detect ROIs of breast tumor ultrasound images effectively and may have the potential applications in the breast tumor Computer Aided Diagnose(CAD) based on ultrasound images.

关键词

Keywords

references

CHENG H D, SHAN J, JU W, et al.. Automated breast cancer detection and classification using ultrasound images: A survey[J]. Pattern Recognition,2010,43(1):299-317.[2] 汪源源, 沈嘉琳. 基于形态特征判别超声图像中乳腺肿瘤的良恶性[J]. 光学精密工程,2006,14(2):333-340. WANG Y Y, SHEN J L. Breast tumor classification based on shape features of ultrasonic images[J]. Opt. Precision Eng.,2006,14(2):333-340.(in Chinese)[3] GUO Y H, CHENG H D, HUANG J H, et al.. Breast ultrasound image enhancement using fuzzy logic[J]. Ultrasound in Med. & Biol., 2006,32(2):237-247.[4] SEHGAL C M, WEINSTEIN S P, ARGER P H, et al.. A review of breast ultrasound[J]. J.Mammary Gland Biol.Neoplasia,2006,11(2):113-123.[5] ANANT M, DIMITRIS N. Metaxas combining low-, high-level and empirical domain knowledge for automated segmentation of ultrasonic breast lesions[J]. IEEE Transactions on Medical Imaging, 2003,22(2):155-169.[6] 马义德, 李廉, 绽琨,等. 脉冲耦合神经网络与数字图像处理[M]. 北京: 科学出版社, 2008:45-67. MA Y D, LI L, ZHAN K, et al.. Pulse Coupled Neural Network and Digital Image Processing[M].Beijing: Science Press,2008:45-67.(in Chinese)[7] SZEKELY G, LINDBLAD T. Parameter adaptation in a simplified pulse-coupled neural network . Proceedings of SPIE Workshop on Virtual Intelligence/Dynamic Neural Networks, Stockholm:SPIE, 1999,3728:278-285.[8] WU C M. Fuzzy mutual information and its application in image segmentation[J]. Computer Engineering,2008,34(7):218-220.[9] 武治国, 王延杰, 李桂菊. 应用小波变换的自适应脉冲耦合神经网络在图像融合中的应用[J]. 光学精密工程,2010,18(3):708-715. WU ZH G, WANG Y J, LI G J. Application of adaptive PCNN based on wavelet transform to image fusion[J]. Opt. Precision Eng.,2010,18(3):708-715.(in Chinese)[10] JOHN L J, MARY L P. PCNN models and applications[J]. IEEE Transactions on Neural Networks,1999,10(3):480-498.[11] LIU B, CHENG H D, HUANG J H, et al.. Automated segmentation of ultrasonic breast lesions using statistical texture classification and active contour based on probability distance[J]. Ultrasound in Medicine and Biology, 2009,35(8):1309-1324.[12] SHI J, XIAO Z, ZHOU S. Automatic segmentation of breast tumor in ultrasound image with simplified PCNN and improved fuzzy mutual information . Visual Communications and Image Processing 2010 Proc. Huangshan, China: SPIE, 2010,7744:241-245.[13] 苏燕妮, 汪源源. 乳腺肿瘤超声图像中感兴趣区域的自动检测[J]. 中国生物医学工程学报,2010,29(2):178-184. SU Y N, WANG Y Y. Automatic detection of the region of interest from breast tumor ultrasound images[J]. Chinese Journal of Biomedical Engineering, 2010,29(2):178-184.(in Chinese)

浏览量

337

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

结合粒子群优化和综合评价的脉冲耦合神经网络图像自动分割

改进的各向异性扩散方程的超声图像滤波方法

基于同质区域自动选取的各向异性扩散超声图像去噪

基于活跃度的脉冲耦合神经网络图像分割

用简化脉冲耦合神经网络实现交通标志图像的类Euclidean距离变换类内特征提取