本文選題：肺結(jié)節(jié)　切入點(diǎn)：CT圖像　出處：《太原理工大學(xué)》2017年碩士論文　論文類(lèi)型：學(xué)位論文

【摘要】：近日,霧霾再次出現(xiàn)并嚴(yán)重影響著人們的生活,同時(shí)危害著人們的身體健康。在霧霾、PM2.5等大氣污染的影響下,全國(guó)乃至全球范圍內(nèi)肺癌患者的數(shù)量呈現(xiàn)出指數(shù)增長(zhǎng)趨勢(shì)。肺癌中病灶體稱(chēng)之為肺結(jié)節(jié),它具有體積較小、形態(tài)各異、多與胸腔內(nèi)膜粘連等特點(diǎn),同時(shí)又有支氣管、血管等干擾,導(dǎo)致對(duì)肺癌的早期診斷具有一定的難度。同時(shí),肺結(jié)節(jié)在肺部所占面積較小,醫(yī)師僅用肉眼對(duì)CT圖像觀察,根據(jù)已有知識(shí)和自身經(jīng)驗(yàn)找出病灶體肺結(jié)節(jié)并對(duì)其進(jìn)行良惡性判斷容易造成誤診或者漏診。在低劑量CT薄層掃描的技術(shù)廣泛應(yīng)用后,影像數(shù)據(jù)爆炸式增長(zhǎng)與人工診斷力量嚴(yán)重不足的矛盾,大數(shù)據(jù)和數(shù)據(jù)分析技術(shù)發(fā)展的不協(xié)調(diào),都有可能導(dǎo)致肺癌診斷準(zhǔn)確率的降低。隨著計(jì)算機(jī)技術(shù)的發(fā)展和應(yīng)用,在眾多大型醫(yī)院中,醫(yī)師都借助于計(jì)算機(jī)對(duì)肺癌進(jìn)行輔助診斷。在較為完善的計(jì)算機(jī)輔助診斷CAD中,通過(guò)圖像預(yù)處理、分割、特征提取和特征選擇等步驟,最終實(shí)現(xiàn)對(duì)肺結(jié)節(jié)的分類(lèi)。提高肺結(jié)節(jié)良惡性分類(lèi)的準(zhǔn)確率是最終的目的,而特征提取是關(guān)鍵的步驟。本文在對(duì)國(guó)內(nèi)外研究現(xiàn)狀進(jìn)行研究后,提出了存在的問(wèn)題還解決辦法。針對(duì)大數(shù)據(jù)的CT圖像,建立基于卷積神經(jīng)網(wǎng)絡(luò)的肺結(jié)節(jié)良惡性自動(dòng)診斷模型,主要的研究工作包括以下幾個(gè)方面:1.針對(duì)計(jì)算機(jī)輔助診斷系統(tǒng)中復(fù)雜的算法和人為干擾產(chǎn)生的不可抗的因素,本文在利用區(qū)域生長(zhǎng)對(duì)CT圖像進(jìn)行簡(jiǎn)單預(yù)處理后得到肺實(shí)質(zhì)圖像,通過(guò)雙線性差值存儲(chǔ)為樣本。利用樣本對(duì)自定義的卷積神經(jīng)網(wǎng)絡(luò)模型進(jìn)行訓(xùn)練并達(dá)到對(duì)肺結(jié)節(jié)良惡性診斷的目的。此方法可以在避免特征提取等復(fù)雜算法的基礎(chǔ)上,提高肺結(jié)節(jié)分類(lèi)準(zhǔn)確性和分類(lèi)速度。2.特征作為肺結(jié)節(jié)主要的分類(lèi)前提,提取特征是必不可少的步驟。在傳統(tǒng)方法中,特征提取方法都是根據(jù)經(jīng)驗(yàn)進(jìn)行人為設(shè)定,包括灰度、形狀以及紋理等底層特征,但是這些特定的特征有一定的局限性。本文直接將圖像作為原始樣本輸入到卷積神經(jīng)網(wǎng)絡(luò)內(nèi),通過(guò)隱含層的自主學(xué)習(xí)提取到關(guān)鍵特征。已有的方法中僅將最后一層輸出作為特征,忽略了隱含層的特征。由于每個(gè)特征的貢獻(xiàn)率不同,且經(jīng)過(guò)多層學(xué)習(xí)可能在最后一層丟失,因此本文將每層特征經(jīng)過(guò)PCA降維得到最終的融合特征。雖然融合特征無(wú)法有確切的描述,但是通過(guò)分類(lèi)器可以得到較為準(zhǔn)確的分類(lèi)結(jié)果。本文通過(guò)搭建卷積神經(jīng)網(wǎng)絡(luò)模型,在大數(shù)據(jù)的樣本的試驗(yàn)認(rèn)證下,在降低了算法的復(fù)雜性的同時(shí)提高了整體肺結(jié)節(jié)的檢出率,降低了誤診率和漏診率。這為醫(yī)師的診斷提供了更為準(zhǔn)確、有效和方便的方法,對(duì)肺癌的早期診斷和治療有著積極作用。
[Abstract]:Recently, haze has reappeared and seriously affected people's lives and their health. Under the influence of air pollution such as haze and PM2.5, The number of lung cancer patients in the whole country and the whole world shows an exponential increasing trend. The focus body of lung cancer is called pulmonary nodule. It has the characteristics of small size, different shape, more adhesion with thoracic endomembranium, and also has bronchus. The early diagnosis of lung cancer is difficult due to the interference of blood vessels. At the same time, the area of pulmonary nodules in the lung is small, so doctors only use the naked eye to observe CT images. It is easy to misdiagnose or miss diagnosis of pulmonary nodules based on their own knowledge and experience. The contradiction between the explosive growth of image data and the serious shortage of artificial diagnostic power, the incoordination between big data and the development of data analysis technology, may lead to a decrease in the diagnostic accuracy of lung cancer. With the development and application of computer technology, In many large hospitals, doctors help diagnose lung cancer with the aid of computer. In CAD, image preprocessing, segmentation, feature extraction and feature selection are used. Finally, the classification of pulmonary nodules is realized. Improving the accuracy of benign and malignant classification of pulmonary nodules is the ultimate goal, and feature extraction is the key step. According to big data's CT image, the automatic diagnosis model of benign and malignant pulmonary nodules based on convolution neural network is established. The main research work includes the following aspects: 1. In view of the complex algorithm and the indelible factors caused by human interference in the computer-aided diagnosis system, the lung parenchyma image is obtained by using the region growth to preprocess the CT image. The bilinear difference is stored as the sample. The self-defined convolution neural network model is trained by the sample and the diagnosis of benign and malignant pulmonary nodules is achieved. This method can avoid complex algorithms such as feature extraction and so on. Improve the accuracy and speed of classification of pulmonary nodules. 2. Feature is the main premise of classification of pulmonary nodules, feature extraction is an essential step. In traditional methods, feature extraction methods are artificially set according to experience, including gray level. In this paper, the image is directly input into the convolutional neural network as the original sample. The key features are extracted by autonomous learning of the hidden layer. In the existing methods, only the output of the last layer is taken as the feature, while the feature of the hidden layer is ignored. Because the contribution rate of each feature is different, and the multi-layer learning may be lost in the last layer, Therefore, the final fusion feature can be obtained by reducing the dimension of each layer by PCA. Although the fusion feature can not be described exactly, the classification result can be obtained by classifier. In this paper, a convolution neural network model is built. Under the experimental verification of big data's sample, the complexity of the algorithm is reduced and the detection rate of global pulmonary nodules is increased, and the misdiagnosis rate and missed diagnosis rate are reduced. This provides a more accurate, effective and convenient method for the diagnosis of doctors. It plays an active role in the early diagnosis and treatment of lung cancer.
【學(xué)位授予單位】：太原理工大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：R734.2;TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 王昊;彭博;陳琴;楊燕;;基于多尺度融合的甲狀腺結(jié)節(jié)圖像特征提取[J];數(shù)據(jù)采集與處理;2016年05期

，

本文編號(hào)：1593763