發(fā)布時間：2018-03-31 01:22

  本文選題：白細(xì)胞　切入點：分類　出處：《深圳大學(xué)》2017年碩士論文

【摘要】：在臨床上,白細(xì)胞分類識別是血液檢驗的一項重要內(nèi)容,準(zhǔn)確、快速的對白細(xì)胞進(jìn)行分類是醫(yī)療領(lǐng)域一項重要的研究。目前,臨床上對白細(xì)胞的檢驗的方法是血細(xì)胞分析儀和人工鏡檢,即先用血細(xì)胞分析儀對樣本進(jìn)行篩查,如果發(fā)現(xiàn)異常樣本,則進(jìn)一步用顯微鏡肉眼觀察,確定最終結(jié)果。人工鏡檢是白細(xì)胞分類的金標(biāo)準(zhǔn),準(zhǔn)確度能夠達(dá)到95%以上。但是人工鏡檢效率低,分類速度慢,準(zhǔn)確度受檢驗人員經(jīng)驗和狀態(tài)的影響。近幾年來,深度學(xué)習(xí)在圖像識別領(lǐng)域取得重大突破,利用這種方法對白細(xì)胞圖像識別成為一種新的研究方向。本文利用深度學(xué)習(xí)的方法設(shè)計了一種白細(xì)胞自動分類系統(tǒng),整個系統(tǒng)包括從血涂片制作到圖像采集分割以及最后的識別分類整個過程。本文完成的主要工作可以概括為以下幾點:1)利用顯微鏡從血涂片中拍攝大量的血細(xì)胞顯微圖像,經(jīng)過圖像分割算法得到大量的單個白細(xì)胞圖像。利用分割得到的白細(xì)胞圖像,建立新的白細(xì)胞數(shù)據(jù)庫,該數(shù)據(jù)庫總共包含四個數(shù)據(jù)集,Train、Train*、Test和Test*。每一個數(shù)據(jù)集都包含五種類型的白細(xì)胞圖像,中性粒細(xì)胞、嗜酸性粒細(xì)胞、嗜堿性粒細(xì)胞、淋巴細(xì)胞和單核細(xì)胞。2)設(shè)計了一種新的深度學(xué)習(xí)網(wǎng)絡(luò)結(jié)構(gòu),包括兩層卷積層、兩層下采樣層、一層全連接層。我們利用Train數(shù)據(jù)集去訓(xùn)練該網(wǎng)絡(luò),并利用Test數(shù)據(jù)集去驗證網(wǎng)絡(luò)模型的性能,白細(xì)胞圖像平均識別率為98.58%。3)通過調(diào)整網(wǎng)絡(luò)參數(shù)和樣本數(shù)量優(yōu)化系統(tǒng)結(jié)構(gòu)。優(yōu)化后的網(wǎng)絡(luò)包含五層卷積層、五層下采樣層和一個全連接層,每個卷積層和下采樣層都包含39個特征映射。優(yōu)化之后,白細(xì)胞圖像平均識別準(zhǔn)確率為99.27%。4)利用交叉驗證的方法評估網(wǎng)絡(luò)模型的性能。我們將Train*和Test*數(shù)據(jù)集中的所有白細(xì)胞圖像分成10等份,9份用作網(wǎng)絡(luò)訓(xùn)練,1份用作網(wǎng)絡(luò)驗證。按照這種方法,我們得到10個不同的白細(xì)胞數(shù)據(jù)庫。我們利用白細(xì)胞數(shù)據(jù)庫得到10組白細(xì)胞識別準(zhǔn)確度,我們?nèi)∑淦骄�值作為系統(tǒng)性能的標(biāo)準(zhǔn)。本文提出了一種基于深度學(xué)習(xí)的白細(xì)胞自動分類方法,并與傳統(tǒng)的白細(xì)胞圖像自動識別方法進(jìn)行了對比。本文提出的方法對中性粒細(xì)胞、嗜酸性粒細(xì)胞、嗜堿性粒細(xì)胞、淋巴細(xì)胞和單核細(xì)胞等五種白細(xì)胞的識別率分別為99.70%、99.63%、99.78%、99.49%、99.62%,平均識別率為99.64%。
[Abstract]:In clinic, leukocyte classification and recognition is an important content of blood test. Accurate and fast classification of white blood cells is an important research in medical field. At present, The methods of clinical examination of white blood cells are hematology analyzer and artificial microscope examination. That is to say, the samples are screened by blood cell analyzer first, and if abnormal samples are found, they are further observed with the naked eye of a microscope. To determine the final results. Artificial microscopy is the gold standard for the classification of white blood cells, and the accuracy can reach more than 95%. But the efficiency of artificial microscopy is low, the speed of classification is slow, and the accuracy is affected by the experience and state of the examiners. In recent years, Depth learning has made a great breakthrough in the field of image recognition. This method has become a new research direction for white blood cell image recognition. In this paper, a white blood cell automatic classification system is designed by using the method of depth learning. The whole system includes the whole process from blood smear making to image acquisition and segmentation, and the final recognition and classification. The main work accomplished in this paper can be summarized as follows: 1) taking a large number of blood cell microscopic images from blood smears by microscope. A large number of single white blood cell images are obtained by image segmentation algorithm, and a new white blood cell database is established by using the white blood cell image obtained by segmentation. The database contains a total of four data sets, Trainberg Trainberg Test and Test.Each dataset contains five types of white blood cell images, neutrophils, eosinophils, basophil, neutrophils, eosinophils, eosinophils, eosinophils, and eosinophils. Lymphocyte and monocyte. 2) designed a new deep learning network structure, which consists of two layers of convolution layer, two layers of lower sampling layer, one layer of full connection layer, and we use the Train data set to train the network. Using the Test data set to verify the performance of the network model, the average recognition rate of leukocyte image is 98.58. 3) by adjusting the network parameters and the number of samples, the system structure is optimized. The optimized network consists of five convolution layers. Five lower sampling layers and a fully connected layer, each convolution layer and lower sampling layer contain 39 feature maps. The average accuracy of leukocyte image recognition is 99.27.4) the performance of the network model is evaluated by cross-validation. We divide all white blood cell images in the Train* and Test* datasets into 10 equal parts and 9 white blood cell images for network training and 1 for network training. Validation. According to this method, We got 10 different white blood cell databases. We used the white blood cell database to get 10 sets of leukocyte recognition accuracy. This paper presents an automatic classification method for white blood cells based on deep learning, and compares it with the traditional method of automatic recognition of white blood cell images. The method proposed in this paper is applied to neutrophilic granulocytes. The recognition rates of eosinophil, basophil, lymphocyte and monocyte were 99.70 and 99.78 respectively. The average recognition rate was 99.64.
【學(xué)位授予單位】：深圳大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：R446.1;TP18;TP391.41

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 傅蓉;;免疫組化彩色細(xì)胞圖像自動分割的研究[J];中國醫(yī)學(xué)物理學(xué)雜志;2008年06期

2 王萌;張留龍;趙運立;;基于顏色矩陣映射的細(xì)胞圖像核、漿提取方法研究[J];中國醫(yī)療設(shè)備;2012年10期

3 湯學(xué)民;林學(xué)，

本文編號：1688537