本文選題：數(shù)據(jù)挖掘 + 人工神經(jīng)網(wǎng)絡(luò)��； 參考：《青島大學(xué)》2017年碩士論文

【摘要】：目的:通過收集的包含多項(xiàng)預(yù)后因素的臨床資料,結(jié)合以人工神經(jīng)網(wǎng)絡(luò)的數(shù)據(jù)挖掘技術(shù),建立預(yù)測(cè)低位直腸癌患者術(shù)后5年生存狀況模型,并和傳統(tǒng)線性統(tǒng)計(jì)相關(guān)分析模型相較,評(píng)價(jià)其性能效果。方法:收集2009年1月~2011年8月在青大附院住院接受手術(shù)治療的186例罹患低位直腸癌的相關(guān)患者的臨床數(shù)據(jù)信息進(jìn)行分析。并隨機(jī)將患者分為兩個(gè)集合:一組共150例,用于數(shù)據(jù)挖掘以建立疾病預(yù)后模型,為訓(xùn)練集;一組共36例,不參加數(shù)據(jù)挖掘,用于模型的性能評(píng)價(jià),為測(cè)試集。此過程中進(jìn)行數(shù)據(jù)挖掘的方式是使用人工神經(jīng)網(wǎng)絡(luò)(ANN)。結(jié)果:T分期、腫瘤最大徑、是否有淋巴結(jié)轉(zhuǎn)移、是否遠(yuǎn)處轉(zhuǎn)移、手術(shù)方式、血清CEA水平、病理類型7項(xiàng)指標(biāo)均與低位直腸癌患者術(shù)后的5年生存相關(guān)(P0.005)。人工神經(jīng)網(wǎng)絡(luò)模型(ANN)對(duì)于患者的5年生存狀況預(yù)測(cè),準(zhǔn)確度為86.11%,敏感度為75.00%,特異度為89.29%,Logistic回歸模型預(yù)測(cè)的結(jié)果顯示,準(zhǔn)確度77.78%,敏感度55.56%,特異度85.19%�？傮w來(lái)看ANN預(yù)測(cè)性能要好于Logistic回歸分析。結(jié)論:ANN數(shù)據(jù)挖掘技術(shù)可從與低位直腸癌患者復(fù)雜的臨床數(shù)據(jù)信息中尋找有意義的預(yù)后相關(guān)的因素,并借助這些因素通過建造預(yù)測(cè)模型,判別患者接受手術(shù)5年后的生存情況。
[Abstract]:Objective: to establish a 5 year survival model for low rectal cancer patients by collecting clinical data including multiple prognostic factors and combining with data mining technology of artificial neural network. Compared with the traditional linear statistical correlation analysis model, the performance of the model is evaluated. Methods: from January 2009 to August 2011, 186 patients with low rectal cancer who were hospitalized in Qingda affiliated Hospital were collected and analyzed. The patients were randomly divided into two sets: one group (150 cases) was used for data mining to establish disease prognosis model, the other group (36 cases) did not participate in data mining, which was used for model performance evaluation and test set. The method of data mining in this process is to use artificial neural network (Ann). Results the tumor size, tumor maximum diameter, lymph node metastasis, distant metastasis, operation mode, serum CEA level and pathological type were all correlated with 5-year survival of patients with low rectal cancer (P 0.005). The Ann model predicted the 5-year survival of patients with accuracy of 86.11, sensitivity of 75.00 and specificity of 89.290.The results of Logistic regression model showed that the accuracy was 77.78, the sensitivity was 55.56, and the specificity was 85.19. Overall, ANN prediction performance is better than Logistic regression analysis. Conclusion\
【學(xué)位授予單位】：青島大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：R735.37

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 閆以聰;;人工神經(jīng)網(wǎng)絡(luò)在醫(yī)學(xué)統(tǒng)計(jì)與信息處理方中的應(yīng)用綜述[J];數(shù)理醫(yī)藥學(xué)雜志;2016年11期

2 張方圓;郁蕓;趙宇;楊坤;胡新華;;人工神經(jīng)網(wǎng)絡(luò)在臨床醫(yī)學(xué)中的應(yīng)用[J];北京生物醫(yī)學(xué)工程;2016年03期

3 衛(wèi)洪波;鄭宗珩;;超低位直腸癌保肛手術(shù)技術(shù)及革新[J];國(guó)際外科學(xué)雜志;2016年01期

4 韓俊毅;陳炳官;;精準(zhǔn)醫(yī)療背景下基因和基因組學(xué)對(duì)外科疾病治療決策的影響[J];腹部外科;2015年04期

5 陳超;郭學(xué)文;唐冬雪;曹延煒;牛海濤;;基于人工神經(jīng)網(wǎng)絡(luò)數(shù)據(jù)挖掘技術(shù)構(gòu)建浸潤(rùn)性膀胱癌預(yù)后模型研究[J];泌尿外科雜志(電子版);2015年02期

6 茆家定;吳佩;楊光;武平;;超低位直腸癌保肛術(shù)的臨床應(yīng)用價(jià)值[J];中華消化外科雜志;2015年06期

7 楊陽(yáng);魏東;;低位直腸癌外科治療新進(jìn)展[J];國(guó)際外科學(xué)雜志;2015年05期

8 黃慶錄;李鴻飛;;腹腔鏡低位直腸癌保肛手術(shù)的研究進(jìn)展[J];中國(guó)微創(chuàng)外科雜志;2015年01期

9 肖峰;丁明躍;尉遲明;;基于運(yùn)動(dòng)量的神經(jīng)網(wǎng)絡(luò)心率預(yù)測(cè)器的設(shè)計(jì)及對(duì)比研究[J];北京生物醫(yī)學(xué)工程;2014年04期

10 傅曉華;洪曉丹;劉石帶;葉毅芳;容穎慈;陳小陸;任斌;;人工神經(jīng)網(wǎng)絡(luò)模型在腎移植患者他克莫司個(gè)體化給藥中的應(yīng)用[J];中國(guó)藥學(xué)雜志;2013年12期

，

本文編號(hào)：1923474