【摘要】：目的:比較BP神經(jīng)網(wǎng)絡(luò)模型和Cox比例風(fēng)險(xiǎn)模型在生存分析中的預(yù)測(cè)性能,進(jìn)一步探討B(tài)P神經(jīng)網(wǎng)絡(luò)模型在生存分析中的應(yīng)用。方法:采用Monte Carlo模擬數(shù)據(jù)集,如不同樣本量、不同刪失比例、不同協(xié)變量間關(guān)系及是否滿足等比例風(fēng)險(xiǎn)假定的理論研究和胃癌根治術(shù)患者預(yù)后預(yù)測(cè)的實(shí)例分析,分別建立BP神經(jīng)網(wǎng)絡(luò)模型和Cox比例風(fēng)險(xiǎn)模型,最終使用一致性指數(shù)C對(duì)其預(yù)測(cè)性能進(jìn)行比較。結(jié)果:當(dāng)樣本量為100、刪失比例為60%、80%及樣本量為300、刪失比例為80%時(shí),BP神經(jīng)網(wǎng)絡(luò)模型的預(yù)測(cè)性能高于Cox比例風(fēng)險(xiǎn)模型(P0.05)。協(xié)變量不滿足等比例風(fēng)險(xiǎn)假定、協(xié)變量間存在三維交互作用和非線性關(guān)系時(shí),BP神經(jīng)網(wǎng)絡(luò)模型的預(yù)測(cè)性能較Cox比例風(fēng)險(xiǎn)模型好(P0.05)。實(shí)例研究中發(fā)現(xiàn),BP神經(jīng)網(wǎng)絡(luò)模型預(yù)測(cè)的一致性指數(shù)C(0.835)高于Cox比例風(fēng)險(xiǎn)模型(t配對(duì)=4.311,P0.001)。結(jié)論:BP神經(jīng)網(wǎng)絡(luò)模型在生存分析的應(yīng)用中對(duì)樣本刪失比例、是否滿足PH假定、協(xié)變量間復(fù)雜交互作用和非線性關(guān)系具有非特異性,對(duì)資料限制較少,且預(yù)測(cè)一致性高,值得在生存分析中進(jìn)一步推廣應(yīng)用。
[Abstract]:Aim: to compare the predictive performance of BP neural network model and Cox proportional hazard model in survival analysis, and further discuss the application of BP neural network model in survival analysis. Methods: Monte Carlo was used to simulate the data sets, such as different sample size, different deletion ratio, the relationship between different covariables and whether equal proportional hazard hypothesis was satisfied, and the case analysis of prognosis prediction of patients undergoing radical operation of gastric cancer was carried out. The BP neural network model and the Cox proportional hazard model are established respectively. Finally, the consistency index C is used to compare the prediction performance. Results: when the sample size was 100, the deletion ratio was 60%, 80% and sample size was 300, and the deletion ratio was 80%, the prediction performance of BP neural network model was higher than that of Cox proportional hazard model (P0.05). The prediction performance of the BP neural network model is better than that of the Cox proportional hazard model when there are three-dimensional interaction and non-linear relationship between the covariables which do not satisfy the assumption of equal proportional hazard (P0.05). In the case study, it is found that the consistency index C (0.835) predicted by the BP neural network model is higher than the Cox proportional hazard model (t pairing = 4.311, P0.001). Conclusion: in the application of BP neural network model to survival analysis, the complexity of interaction and non-linear relationship between covariates is non-specific, and the prediction consistency is high, and the ratio of sample deletion is satisfied with the PH hypothesis. It is worth further popularizing and applying in the survival analysis.
【作者單位】： 徐州醫(yī)學(xué)院公共衛(wèi)生學(xué)院流行病與衛(wèi)生統(tǒng)計(jì)學(xué)教研室;
【基金】：江蘇省科技廳資助項(xiàng)目BE2011647
【分類號(hào)】：R741;R-332

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 高蔚,施侶元;人工神經(jīng)網(wǎng)絡(luò)流行病學(xué)應(yīng)用進(jìn)展[J];中華預(yù)防醫(yī)學(xué)雜志;2000年06期

相關(guān)博士學(xué)位論文 前1條

1 錢俊;生存分析中刪失數(shù)據(jù)比例對(duì)Cox回歸模型影響的研究[D];南方醫(yī)科大學(xué);2009年

【共引文獻(xiàn)】

相關(guān)期刊論文 前8條

1 周其宏;馮曉明;汪銀;汪攀文;寧玲;王小紅;;皖南山區(qū)流行性腮腺炎發(fā)病趨勢(shì)的智能預(yù)測(cè)模型[J];中華疾病控制雜志;2010年08期

2 蘇錦霞;張藝贏;田麗娜;;Ⅰ型逐階區(qū)間刪失Weibull數(shù)據(jù)的統(tǒng)計(jì)分析[J];蘭州大學(xué)學(xué)報(bào)(自然科學(xué)版);2011年05期

3 王立芹;唐龍妹;閆麗娜;;醫(yī)學(xué)科研中常用的幾種多重回歸模型[J];臨床薈萃;2013年05期

4 張僅;周瑞祥;尚柏林;張忠;;基于環(huán)境因子的飛機(jī)液壓導(dǎo)管壽命分析[J];火力與指揮控制;2014年11期

5 任宏;人工神經(jīng)網(wǎng)絡(luò)及其在預(yù)防醫(yī)學(xué)領(lǐng)域的應(yīng)用[J];上海預(yù)防醫(yī)學(xué)雜志;2003年01期

6 高蔚;聶紹發(fā);施侶元;;神經(jīng)網(wǎng)絡(luò)在生存分析中的應(yīng)用進(jìn)展[J];中國(guó)衛(wèi)生統(tǒng)計(jì);2006年04期

7 徐俊芳;周曉農(nóng);;人工神經(jīng)網(wǎng)絡(luò)在傳染病研究中的應(yīng)用[J];中國(guó)寄生蟲(chóng)學(xué)與寄生蟲(chóng)病雜志;2011年01期

8 俞剛;鄭q;葉盛;;基于模糊C均值聚類的兒科機(jī)械通氣撤機(jī)時(shí)機(jī)研究[J];中國(guó)生物醫(yī)學(xué)工程學(xué)報(bào);2014年05期

相關(guān)會(huì)議論文 前1條

1 周其宏;馮曉明;汪銀;汪攀文;寧玲;王小紅;;皖南山區(qū)流行性腮腺炎發(fā)病趨勢(shì)的智能預(yù)測(cè)模型[A];華東地區(qū)第十次流行病學(xué)學(xué)術(shù)會(huì)議暨華東地區(qū)流行病學(xué)學(xué)術(shù)會(huì)議20周年慶典論文集[C];2010年

相關(guān)博士學(xué)位論文 前4條

1 張波;影像分析技術(shù)在疾病監(jiān)測(cè)與診斷中的應(yīng)用[D];第四軍醫(yī)大學(xué);2005年

2 范p，

本文編號(hào)：2436896