新疆高发病食管癌图像的特征提取及分类
|
摘要
目的结合灰度共生矩阵和小波变换的纹理分析方法提取新疆哈萨克族高发病食管癌X射线钡剂造影图像的特征,旨在为放射科医生的诊断决策提供具有实际参考价值的辅助信息,提高食管癌诊断的准确率和效率。方法选取2种中晚期食管癌——蕈伞型和缩窄型,以及正常食管图像各100张,利用基于灰度共生矩阵的纹理特征提取方法分别提取食管癌X射线图像的角二阶矩、熵、惯性矩、逆差矩及相关性的方差作为纹理特征,同时使用小波变换对食管癌X射线图像进行二层小波分解,获取其高频子图,并提取高频子图的能量特征作为纹理特征。然后,使用C4.5决策树算法构造一个分类器,对正常食管和中晚期食管癌图像进行分类研究。结果共计提取11维特征,利用单一特征算法进行分类,灰度共生矩阵法分类准确率为64.66%,小波变换法分类准确率为77%。而综合的灰度共生矩阵和小波变换法的分类准确率为81.67%,更适用于正常食管和中晚期食管癌的分类。结论本研究将灰度共生矩阵、小波变换算法与决策树C4.5相结合,对正常食管与蕈伞型和缩窄型食管癌进行特征提取及分析,结果表明本算法分类准确率较高,为开发食管癌的计算机辅助诊断系统奠定了基础。
Objective In this paper,combining gray level co-occurrence matrix( GLCM) with wavelet transform of texture arithmetic method extracted X-ray image texture of esophageal cancer with high morbidity in Xinjiang. This method provided auxiliary information with reference value for radiologist and enhanced accuracy rate and efficiency of esophageal cancer. Methods Two types of middle and terminal esophageal cancer:fungating type esophageal cancer and constrictive esophageal cancer were selected in the experiment,and 100 images of the two types of esophageal cancers and normal esophagus were selected,respectively. We used GLCM texture feature method to extract angular second moment,entropy,contrast,correlation and inverse difference moment of variance respectively for X-ray images of esophageal cancer. Meanwhile,we employed wavelet transform to process two-dimensional discrete wavelet decomposition at second level,obtained its highfrequency content images,and extracted the energy features of high-frequency content images as texture features. The C4. 5 decision tree was employed as a classifier. Results Eleven features were extracted by GLCM and wavelet transform methods. The experimental results showed that using single feature classification,theaccuracy rate of GLCM classification and wavelet transform classification reached to 64. 66% and 77%,respectively. The accuracy rate of the comprehensive of GLCM and wavelet transform method was 81. 67%,more suitable for the classification of normal esophagus and advanced esophageal cancer. Conclusions This method achieved high classification performance and lay the foundation of the computer-aided diagnosis system of Kazakh esophageal cancer in Xinjiang.
Objective In this paper,combining gray level co-occurrence matrix( GLCM) with wavelet transform of texture arithmetic method extracted X-ray image texture of esophageal cancer with high morbidity in Xinjiang. This method provided auxiliary information with reference value for radiologist and enhanced accuracy rate and efficiency of esophageal cancer. Methods Two types of middle and terminal esophageal cancer:fungating type esophageal cancer and constrictive esophageal cancer were selected in the experiment,and 100 images of the two types of esophageal cancers and normal esophagus were selected,respectively. We used GLCM texture feature method to extract angular second moment,entropy,contrast,correlation and inverse difference moment of variance respectively for X-ray images of esophageal cancer. Meanwhile,we employed wavelet transform to process two-dimensional discrete wavelet decomposition at second level,obtained its highfrequency content images,and extracted the energy features of high-frequency content images as texture features. The C4. 5 decision tree was employed as a classifier. Results Eleven features were extracted by GLCM and wavelet transform methods. The experimental results showed that using single feature classification,theaccuracy rate of GLCM classification and wavelet transform classification reached to 64. 66% and 77%,respectively. The accuracy rate of the comprehensive of GLCM and wavelet transform method was 81. 67%,more suitable for the classification of normal esophagus and advanced esophageal cancer. Conclusions This method achieved high classification performance and lay the foundation of the computer-aided diagnosis system of Kazakh esophageal cancer in Xinjiang.
引文
[1]李丹.食管癌研究进展[J].吉林中医药,2012,32(9):970-972.Li Dan.The research progress of esophageal cancer[J].Jilin Journal of Traditional Chinese Medicine,2012,32(9):970-972.
[2]郭辉,丁剑冰,孙伟,等.新疆哈萨克族食管癌相关基因研究新进展[J].基础医学与临床,2010,30(4):428-430.Guo Hui,Ding Jianbing,Sun Wei,et al.Research progress of related genes in Xinjiang Kazakh’s esophageal cancer[J].Basic&Clinical Medicine,2010,30(4):428-430.
[3]刘博.基于小波变换特征的医学图像分类研究[D].镇江:江苏大学,2010.Liu Bo.Classification of medical images using wavelet-based transform feature[D].Zhenjiang:Jiangsu University,2010.
[4]孙旭辉,田启川,李临生,等.面向CT影像的肺结节计算机辅助诊断算法[J].计算机测量与控制,2011,19(2):295-302.Sun Xuhui,Tian Qichuan,Li Linsheng,et al.Computer-aided detection algorithm for pulmonary nodule based on CT images[J].Computer Measurement&Control,2011,19(2):295-302.
[5]童振,蒲立新,曲建明.基于乳腺X线成像的计算机辅助诊断技术研究进展[J].中国数字医学,2011,6(2):98-101.Tong Zhen,Pu Lixin,Qu Jianming.Computer-aided diagnosis of breast cancer with mammography recent advances[J].China Digital Medicine,2011,6(2):98-101.
[6]程智辉,陈将宏.基于概率神经网络的乳腺癌计算机辅助诊断[J].计算机仿真,2012,29(9):166-169.Cheng Zhihui,Chen Jianghong.Computer-aided diagnosis of breast cancer based on probabilistic neural networks[J].Computer Simulation,2012,29(9):166-169.
[7]李玲,余后强.基于CT图像的肝癌计算机辅助诊断[J].甘肃科学学报,2012,24(1):60-64.Li Ling,Yu Houqiang.Computer-aided diagnosis of liver cancer based on CT images[J].Journal of Gansu Science,2012,24(1):60-64.
[8]胡玉娜,冯开颜,张志勇,等.基于形态特征的肺癌CT图像计算机辅助诊断初探[J].实用放射学杂志2012,28(2):293-296.Hu Yuna,Feng Kaiyan,Zhang Zhiyong,et al.Computer-aided of lung cancer imaged by CT:initial test result[J].Journal of Practical Radiology,2012,28(2):293-296.
[9]Armato SR,Giger ML,Moran CJ,et al.Computerized detection of pilmonary nodules on CT scans.Radiographics,1999,19(5):1303-1311.
[10]陈秋霞,向军,刘奇,等.乳腺肿瘤超声图像纹理特征分析[J].激光杂志,2014,35(3):65-69.Chen Qiuxia,Xiang Jun,Liu Qi.Textural feature analysis for ultrasound breast tumor images[J].Laser Journal,2014,35(3):65-69.
[11]杨国城,卢凤梅,李美玲.基于灰度共生矩阵的肝脏纹理特征提取[J].医药前沿,2012,2(17):105-106.Yang Guocheng,Lu Fengmei,Li Meiling.Feature extraction of liver texture based on the co-occurrence matrix[J].Journal of Frontiers of Medicine,2012,2(17):105-106.
[12]付燕,聂亚娜,靳玉萍.基于SVM的肝脏B超图像纹理分类[J].计算机应用与软件,2012,29(12):291-294.Fu Yan,Nie Yana,Jin Yuping.Texture classification of B ultrasound images of liver based on SVM[J].Computer Applications and Software,2012,29(12):291-294.
[13]聂会星,梁坤,徐枞巍.基于小波变换和支持向量机的人脸识别研究[J].合肥工业大学学报:自然科学版,2011,34(2):208-211.Nie Huixing,Liang Kun,Xu Zongwei.Face recognition based on wavelet transform and support vector machine[J].Journal of Hefei University of Technology:Science and Technology,2011,34(2):208-211.
[14]张慧,迟庆云,刘彩霞.基于灰度共生矩阵的肝癌B超纹理特征决策树诊断分析[J].中国医药指南,2015,13(25):2-3.Zhang Hui,Chi Qingyun,Liu Caixia.Decision tree diagnosis analysis of liver B ultrasonic feature[J].Guide of China Medicine,2015,13(25):2-3.
[15]张炎.影像学检查对食管肿瘤的价值[J].中国医药科学,2011,1(8):5-9.Zhang Yan.Imageological examination of value for esophageal tumour[J].China Medicine and Pharmacy,2011,1(8):5-9.
[16]木拉提·哈米提,杨芳,孔德伟,等.新疆高发病哈萨克族食管癌图像纹理特征的分类研究[J].新疆医科大学学报,2014,37(3):273-276.Murat Hamit,Yang Fang,Kong Dewei,et al.Classification on Xinjiang high morbidity of Kazak esophageal disease based on texture feature[J].Journal of Xinjiang Medical University,2014,37(3):273-276.
[17]杨韬,夏建国,缪锦林,等.基于灰度共生矩阵的乳腺钼靶图像分析[J].中国医学影像学杂志,2012,20(4):306-309.Yang Tao,Xia Jianguo,Miao Jinlin,et al.Analysis of gray level co-occurrence matrix mammogram[J].Chinese Journal of Medical Imaging,2012,20(40):306-309.
[18]王红君,施楠,赵辉,等.改进中值滤波方法的图像预处理技术[J].计算机系统应用,2015,24(5):237-240.Wang Hongjun,Shi Nan,Zhao Hui,et al.Image preprocessing technology based on improved median filter[J].Computer Systems&Application,2015,24(5):237-240.
[19]彭丹.小波分析概述及其应用研究[J].装备制造技术,2014(6):145-153.Peng Dan.Wavelet analysis overview and its application[J].Equipment Manufacturing Technology,2014(6):145-153.
[20]Nanthagopal AP,Rajamony RS.Automatic classification of brain computed tomography images using wavelet-based statistical texture features[J].The Visualization Society of Japan,2012,7:1-10.
[21]Amin HU,Malik AS,Ahmad RF,et al.Feature extraction and classification for EEG signals using wavelet transform and machine learning techniques[J].Australasian Physical&Engineering Sciences in Medicine,2015,38(1):139-149.
[22]林森森,姜新国,彭小芳,等.基于灰度共生矩阵的乳腺肿瘤纹理特征提取[J].徐州医学院学报,2013,33(3):196-198.Lin Sensen,Jiang Xinguo,Peng Xiaofang,et al.Extraction of breast tumor texture features based on gray level co-occurrence matrix[J].Acta Academiae Medicinae Xuzhou,2013,33(3):196-198.
[23]彭庆涛,吴水才,高宏建,等.基于小波分析和灰度纹理特征的乳腺X线图像微钙化点区域的提取[J].北京生物医学工程,2015,34(5):462-468.Peng Qingtao,Wu Shuicai,Gao Hongjian,et al.Extraction of the regions of microcalcification in breast X-ray image based on wavelet analysis and image texture feature[J].Beijing Biomedical Engineering,2015,34(5):462-468.
[24]Haralick RM.Statistical and structural approaches to texture[J].Proceedings of the IEEE,1979,67(5):786-840.
[25]Conners RW,Harlow CA.A theoretical comparison of texture algorithms[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1980,2(3):204-222.
[2]郭辉,丁剑冰,孙伟,等.新疆哈萨克族食管癌相关基因研究新进展[J].基础医学与临床,2010,30(4):428-430.Guo Hui,Ding Jianbing,Sun Wei,et al.Research progress of related genes in Xinjiang Kazakh’s esophageal cancer[J].Basic&Clinical Medicine,2010,30(4):428-430.
[3]刘博.基于小波变换特征的医学图像分类研究[D].镇江:江苏大学,2010.Liu Bo.Classification of medical images using wavelet-based transform feature[D].Zhenjiang:Jiangsu University,2010.
[4]孙旭辉,田启川,李临生,等.面向CT影像的肺结节计算机辅助诊断算法[J].计算机测量与控制,2011,19(2):295-302.Sun Xuhui,Tian Qichuan,Li Linsheng,et al.Computer-aided detection algorithm for pulmonary nodule based on CT images[J].Computer Measurement&Control,2011,19(2):295-302.
[5]童振,蒲立新,曲建明.基于乳腺X线成像的计算机辅助诊断技术研究进展[J].中国数字医学,2011,6(2):98-101.Tong Zhen,Pu Lixin,Qu Jianming.Computer-aided diagnosis of breast cancer with mammography recent advances[J].China Digital Medicine,2011,6(2):98-101.
[6]程智辉,陈将宏.基于概率神经网络的乳腺癌计算机辅助诊断[J].计算机仿真,2012,29(9):166-169.Cheng Zhihui,Chen Jianghong.Computer-aided diagnosis of breast cancer based on probabilistic neural networks[J].Computer Simulation,2012,29(9):166-169.
[7]李玲,余后强.基于CT图像的肝癌计算机辅助诊断[J].甘肃科学学报,2012,24(1):60-64.Li Ling,Yu Houqiang.Computer-aided diagnosis of liver cancer based on CT images[J].Journal of Gansu Science,2012,24(1):60-64.
[8]胡玉娜,冯开颜,张志勇,等.基于形态特征的肺癌CT图像计算机辅助诊断初探[J].实用放射学杂志2012,28(2):293-296.Hu Yuna,Feng Kaiyan,Zhang Zhiyong,et al.Computer-aided of lung cancer imaged by CT:initial test result[J].Journal of Practical Radiology,2012,28(2):293-296.
[9]Armato SR,Giger ML,Moran CJ,et al.Computerized detection of pilmonary nodules on CT scans.Radiographics,1999,19(5):1303-1311.
[10]陈秋霞,向军,刘奇,等.乳腺肿瘤超声图像纹理特征分析[J].激光杂志,2014,35(3):65-69.Chen Qiuxia,Xiang Jun,Liu Qi.Textural feature analysis for ultrasound breast tumor images[J].Laser Journal,2014,35(3):65-69.
[11]杨国城,卢凤梅,李美玲.基于灰度共生矩阵的肝脏纹理特征提取[J].医药前沿,2012,2(17):105-106.Yang Guocheng,Lu Fengmei,Li Meiling.Feature extraction of liver texture based on the co-occurrence matrix[J].Journal of Frontiers of Medicine,2012,2(17):105-106.
[12]付燕,聂亚娜,靳玉萍.基于SVM的肝脏B超图像纹理分类[J].计算机应用与软件,2012,29(12):291-294.Fu Yan,Nie Yana,Jin Yuping.Texture classification of B ultrasound images of liver based on SVM[J].Computer Applications and Software,2012,29(12):291-294.
[13]聂会星,梁坤,徐枞巍.基于小波变换和支持向量机的人脸识别研究[J].合肥工业大学学报:自然科学版,2011,34(2):208-211.Nie Huixing,Liang Kun,Xu Zongwei.Face recognition based on wavelet transform and support vector machine[J].Journal of Hefei University of Technology:Science and Technology,2011,34(2):208-211.
[14]张慧,迟庆云,刘彩霞.基于灰度共生矩阵的肝癌B超纹理特征决策树诊断分析[J].中国医药指南,2015,13(25):2-3.Zhang Hui,Chi Qingyun,Liu Caixia.Decision tree diagnosis analysis of liver B ultrasonic feature[J].Guide of China Medicine,2015,13(25):2-3.
[15]张炎.影像学检查对食管肿瘤的价值[J].中国医药科学,2011,1(8):5-9.Zhang Yan.Imageological examination of value for esophageal tumour[J].China Medicine and Pharmacy,2011,1(8):5-9.
[16]木拉提·哈米提,杨芳,孔德伟,等.新疆高发病哈萨克族食管癌图像纹理特征的分类研究[J].新疆医科大学学报,2014,37(3):273-276.Murat Hamit,Yang Fang,Kong Dewei,et al.Classification on Xinjiang high morbidity of Kazak esophageal disease based on texture feature[J].Journal of Xinjiang Medical University,2014,37(3):273-276.
[17]杨韬,夏建国,缪锦林,等.基于灰度共生矩阵的乳腺钼靶图像分析[J].中国医学影像学杂志,2012,20(4):306-309.Yang Tao,Xia Jianguo,Miao Jinlin,et al.Analysis of gray level co-occurrence matrix mammogram[J].Chinese Journal of Medical Imaging,2012,20(40):306-309.
[18]王红君,施楠,赵辉,等.改进中值滤波方法的图像预处理技术[J].计算机系统应用,2015,24(5):237-240.Wang Hongjun,Shi Nan,Zhao Hui,et al.Image preprocessing technology based on improved median filter[J].Computer Systems&Application,2015,24(5):237-240.
[19]彭丹.小波分析概述及其应用研究[J].装备制造技术,2014(6):145-153.Peng Dan.Wavelet analysis overview and its application[J].Equipment Manufacturing Technology,2014(6):145-153.
[20]Nanthagopal AP,Rajamony RS.Automatic classification of brain computed tomography images using wavelet-based statistical texture features[J].The Visualization Society of Japan,2012,7:1-10.
[21]Amin HU,Malik AS,Ahmad RF,et al.Feature extraction and classification for EEG signals using wavelet transform and machine learning techniques[J].Australasian Physical&Engineering Sciences in Medicine,2015,38(1):139-149.
[22]林森森,姜新国,彭小芳,等.基于灰度共生矩阵的乳腺肿瘤纹理特征提取[J].徐州医学院学报,2013,33(3):196-198.Lin Sensen,Jiang Xinguo,Peng Xiaofang,et al.Extraction of breast tumor texture features based on gray level co-occurrence matrix[J].Acta Academiae Medicinae Xuzhou,2013,33(3):196-198.
[23]彭庆涛,吴水才,高宏建,等.基于小波分析和灰度纹理特征的乳腺X线图像微钙化点区域的提取[J].北京生物医学工程,2015,34(5):462-468.Peng Qingtao,Wu Shuicai,Gao Hongjian,et al.Extraction of the regions of microcalcification in breast X-ray image based on wavelet analysis and image texture feature[J].Beijing Biomedical Engineering,2015,34(5):462-468.
[24]Haralick RM.Statistical and structural approaches to texture[J].Proceedings of the IEEE,1979,67(5):786-840.
[25]Conners RW,Harlow CA.A theoretical comparison of texture algorithms[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1980,2(3):204-222.