摘要
针对临床医学中急性淋巴细胞白血病(Acute Lymphoblastic Leukemia,ALL)血液细胞显微图像分类易错、费时等问题,本文提出了一种基于深度学习VGG16卷积神经网络模型的方法去获取医学图像中高纬度的病理信息。该方法首先将样本数据进行预处理,清洗出符合要求的训练集和验证集,其中还用到了超像素的方法用于训练样本的目标区域提取,然后再将预处理好的数据,输入到VGG16卷积神经网络模型中,对其进行训练,最后输入验证集进入模型中进行验证,实验结果表明,该分类方法能有效地完成ALL血液细胞显微图像是否患病的分类。
Aiming at the problem of accurate classification and time consuming in clinical medicine microscopic images of acute lymphoblastic leukemia(ALL), this paper proposed a method based on deep learning VGG16 convolutional neural network model to obtain pathological information at high latitudes in medical images. In the paper, we firstly preprocessed the sample, cleaned out the training set and verification set that met the requirements, and also used the super-pixel method to extract the target area of the training sample. Then we trained VGG16 network by inputting the preprocessed data, and finally the validation set was entered into the model for verification. The experimental result showed that the classification method could effectively complete the classification of ALL blood cell microscopic images.
引文
[1]孙璀.白血病临床诊断中血常规指标综合分析的应用价值评析[J].中国现代药物应用,2016,10(6):94-95.
[2]赵玲玲,陈响亮,徐骁盟,等.白细胞分类计数对急性脑梗死患者的临床预测价值[J].医学研究生学报,2015,28(11):1148-1151.
[3]靳锐玉.急性白血病血常规指标及白细胞分类分析[J].基层医学论坛,2018,22(17):2385-2386.
[4]刘广花,周明.血细胞参数在白血病诊断及预后中的价值研究[J].中国社区医师,2018,34(32):112-113.
[5]战伟.白细胞分类人工镜检与全自动血细胞分析仪检查结果的比较[J].大连大学学报,2018,39(3):62-66.
[6]尹娟.Sysmex XT-1800i五分类血球计数仪白细胞分类报警提示的结果分析[J].吉林医学,2013,34(13):2526.
[7]张会娟.血球计数仪所测的WBC误差分析[J].锦州医学院学报,1999,(3):77.
[8]宋宏伟,季伙燕,王建新,等.白细胞分类计数的不确定度评定[J].检验医学与临床,2017,14(1):27-28.
[9]索荣华,宋毅.MEK血球计数仪检测误差分析[J].医疗设备信息,2005,(4):74-75.
[10]庞春颖,刘记奎,韩立喜.改进FCM和LFP相结合的白细胞图像分类[J].中国图象图形学报,2013,18(5):545-551.
[11]李小舜,曹益平,王亚品.基于mean-shift聚类的高鲁棒性白细胞五分类识别算法[J].生物医学工程学杂志,2018,35(5):761-766.
[12]李少坤.SMO算法与决策树算法在医疗科技应用中的对比研究[J].中国高新科技,2019,(1):127-128.
[13]柳秋云.改进的朴素贝叶斯分类器在医疗诊断中的应用[J].科技创新导报,2008,(31):192.
[14]怀听听,赵建伟,曹飞龙,等.基于综合特征和随机森林的白细胞分类算法[J].中国计量学院学报,2015,26(4):474-479.
[15]Litjens G,Kooi T,Bejnordi BE,et al.A survey on deep learning in medical image analysis[J].Med Image Anal,2017,42(9):60-88.
[16]Tabrizi PR,Rezatofighi SH,Yazdanpanah MJ.Using PCA and LVQneural network for automatic recognition of five types of white blood cells[J].Conf Proc IEEE Eng Med Biol Soc,2010,2010(10):5593-5596.
[17]Manik S,Saini LM,Vadera N.Counting and classification of white blood cell using Artificial Neural Network(ANN)[A].IEEE International Conference on Power Electronics[C].2017.
[18]Habibzadeh M,Jannesari M,Rezaei Z,et al.Automatic white blood cell classification using pre-trained deep learning models:ResNet and Inception[A].Tenth International Conference on Machine Vision(ICMV 2017)[C].2018:1069612.
[19]陈畅,程少杰,李卫滨,等.基于卷积神经网络的外周血白细胞分类[J].中国生物医学工程学报,2018,37(1):17-24.
[20]魏存超.基于卷积神经网络的医学图像分类的研究[D].哈尔滨:哈尔滨工业大学,2017.
[21]韩冬,李其花,蔡巍,等.人工智能在医学影像中的研究与应用[J].大数据,2019,5(1):39-67.
[22]Labati RD,Piuri V,Scotti F.All-IDB:The acute lymphoblastic leukemia image database for image processing[A].2011 18th IEEE International Conference on Image Processing[C].New York:IEEE,2011:2045-2048.
[23]Othman M Z,Mohammed T S,AliA B.Neural network classification of white blood cell using microscopic images[J].IJACSA,2017,8(5):99-104.C