基于遗传算法-支持向量机的兔肝VX2肿瘤光谱鉴别  

Spectral Discrimination of Rabbit Liver VX2 Tumor and Normal Tissue Based on Genetic Algorithm-Support Vector Machine

作　　者：刘晨阳 许黄蓉 段峰[4] 王泰升[1] 卢振武[1] 鱼卫星[3] LIU Chen-yang;XU Huang-rong;DUAN Feng;WANG Tai-sheng;LU Zhen-wu;YU Wei-xing(State Key Laboratory of Applied Optics,Changchun Institute of Optics,Fine Mechanics&Physics,Chinese Academy of Sciences,Changchun 130033,China;University of Chinese Academy of Sciences,Beijing 100049,China;Key Laboratory of Spectral Imaging Technology,Xi’an Institute of Optics and Precision Mechanics,Chinese Academy of Science,Xi’an 710119,China;Department of Interventional Radiology,the General Hospital of Chinese People’s Liberation Army,Beijing 100853,China)

机构地区：[1]中国科学院长春光学精密机械与物理研究所精密仪器与装备研发中心,吉林长春130033 [2]中国科学院大学,北京100049 [3]中国科学院西安光学精密机械研究所,中国科学院光谱成像技术重点实验室,陕西西安710119 [4]中国人民解放军总医院介入放射科,北京100853

出　　处：《光谱学与光谱分析》2021年第10期3123-3128,共6页Spectroscopy and Spectral Analysis

基　　金：国家自然科学基金面上项目(61975231);吉林省重点科技研发项目(20180201045YY);吉林省重点科技研发项目(20180201041YY)资助。

摘　　要：兔肝VX2肿瘤是一种快速生长的肿瘤模型,可以在多种器官如肝、肺、直肠等快速生长,常用于肿瘤研究。采用可见-近红外高光谱技术对四只兔子的兔肝VX2肿瘤和正常组织进行活体和离体的反射光谱检测,然后采用支持向量机分别实现了二分类(正常肝组织和肝VX2肿瘤组织)和四分类(未出血活体正常肝组织、未出血活体VX2肿瘤组织、出血离体正常肝组织和出血离体肝VX2肿瘤组织)。根据其光谱反射曲线的特征,选择了400~1800 nm区间的数据为特征变量。为进一步提高分类准确率,分别采用5折交叉验证和遗传算法对支持向量机的核函数参数g和惩罚因子c进行了优化。其中5折交叉验证优化参数和分类结果为:二分类优化的惩罚参数c为4,核函数参数g为0.1250,其校正集和预测集的准确率都达到了100%;四分类中优化出的参数c为8,g为0.1211,其校正集和预测集的准确率分别达到了99.2424%和93.333%。遗传算法优化参数和结果为:二分类中优化的参数c为0.8456,g为0.0625,其校正集和预测集的准确率同样都达到了100%;四分类中优化的参数c为5.5307,g为0.0685,其校正集和预测集的准确率分别达到了99.2424%和100%。结果显示两种优化方法都取得了很好的效果,遗传算法优化参数对四分类的分类更为精确。为进一步提升算法速度,采用间隔选取变量的方法来不断减少特征变量,最终每隔100 nm谱段选择一个变量,共选择14个谱段作为特征变量。采用遗传算法优化支持向量机参数并对其分类进行了研究,结果表明:二分类和四分类的校正集和预测集结果准确率均为99.2424%,而且运行时间分别为11.4和20.0 s,与选择全波段的运行时间:340.3和491.0 s相比,说明多光谱技术可以进行肝VX2肿瘤组织和正常肝组织的鉴别,且分类准确率可达99%以上,而且运行时间缩短了很多。为未来多光谱技术在未来临床肿瘤诊断中实现�Rabbit liver VX2 tumor is a tumor model that can grow rapidly in various organs,such as liver,lung,rectum,etc.,and is often used in tumor research.In this paper,using high-near-infrared spectrum technology to four rabbits VX2 liver tumor and normal tissue in vivo and in vitro reflection spectrum detection,then respectively the Two categories based on support vector machine(normal liver tissue and liver VX2 tumor tissue)and Four categories(not bleeding living normal liver tissue,not living liver VX2 tumor tissue bleeding,bleeding in vitro normal liver tissue and hemorrhage in vitro liver VX2 tumor tissue).According to its spectral reflection curve characteristics,the data in the range of 400~1800 nm are selected as characteristic variables.In order to further improve the classification accuracy,the kernel parameter g and penalty factor c of the support vector machine was optimized by using a 50 fold cross-validation and genetic algorithm,respectively.The optimization parameters and classification results of the 50-fold cross-validation are as follows:penalty parameter c of the dichotomy optimization is 4,kernel parameter g is 0.1250,and the accuracy of the correction set and prediction set reaches 100%.The optimized parameters c and g are 8 and 0.1211,and the accuracy of the correction set and the prediction set are 99.2424%and 93.333%,respectively.The optimized parameters and results of the genetic algorithm are as follows:the optimized parameters c and g in dichotomy are 0.8456 and 0.0625,respectively,and the accuracy of Two categories,the correction set and the prediction set,is agreed to reach 100%.The optimized parameter C in the Four categories was 5.5307 and g was 0.0685,and the accuracy of the correction set and the prediction set reached 99.2424%and 100%,respectively.The results show that the two optimization methods have achieved good results,and the genetic algorithm is more accurate in the classification of the Four categories.In order to further improve the speed of the algorithm,the method of variabl

关 键 词：兔肝VX2肿瘤 可见-近红外光谱 遗传算法 支持向量机 

分 类 号：O434.3[机械工程—光学工程]