Development and Evaluation of Novel Statistical Methods in Urine Biomarker-Based Hepatocellular Carcinoma Screening.
Development and Evaluation of Novel Statistical Methods in Urine Biomarker-Based Hepatocellular Carcinoma Screening.
复制标题
DOI：
10.1038/s41598-018-21922-9
复制
发表时间：
2018-02-28
影响因子：
4.6
通讯作者：
Su YH
中科院分区：
文献类型：
--
作者：
Wang J;Jain S;Chen D;Song W;Hu CT;Su YH
肝细胞癌是美国增长最快的癌症之一，生存率很低，部分原因是早期发现困难。由于HCC的高度异质性，有人认为需要多种生物标志物来开发一种敏感的HCC筛查试验。本研究应用随机森林（RF），一种机器学习技术，提出了两种新的模型，固定顺序（FS）和两步（TS），与两种常用的统计技术，逻辑回归（LR）和分类回归树（CART）进行比较，结合多种尿液DNA生物标志物进行HCC筛查，使用从137例HCC和431例非HCC（224例肝炎和207例肝硬化）受试者中获得的生物标志物值。通过重复的10倍交叉验证来估计灵敏度、特异性、受试者工作曲线下面积和变异性，以比较模型的准确性和稳健性。结果表明，RF和TS具有更高的精度和稳定性；在交叉验证中，它们的特异性分别达到90%和86%/87%，灵敏度比LR高15%，特异性比LR高10%。讨论了RF和TS开发多种生物标志物的潜力，以及自我训练的可能性，基于云的HCC筛查模型。
Hepatocellular carcinoma is one of the fastest growing cancers in the US and has a low survival rate, partly due to difficulties in early detection. Because of HCC’s high heterogeneity, it has been suggested that multiple biomarkers would be needed to develop a sensitive HCC screening test. This study applied random forest (RF), a machine learning technique, and proposed two novel models, fixed sequential (FS) and two-step (TS), for comparison with two commonly used statistical techniques, logistic regression (LR) and classification and regression trees (CART), in combining multiple urine DNA biomarkers for HCC screening using biomarker values obtained from 137 HCC and 431 non-HCC (224 hepatitis and 207 cirrhosis) subjects. The sensitivity, specificity, area under the receiver operating curve, and variability were estimated through repeated 10-fold cross-validation to compare the models’ performances in accuracy and robustness. We show that RF and TS have higher accuracy and stability; specifically, they reach 90% specificity and 86%/87% sensitivity respectively along with 15% higher sensitivity and 10% higher specificity than LR in cross-validation. The potential of RF and TS to develop a panel of multiple biomarkers and the possibility for self-training, cloud-based models for HCC screening are discussed.
登录
查看更多内容
影响因子：
5.4
作者：
Ozen, Cigdem;Yildiz, Gokhan;Ozturk, Mehmet
通讯作者：
Ozturk, Mehmet
影响因子：
29.4
作者：
El-Serag HB
通讯作者：
El-Serag HB
影响因子：
29.4
作者：
Daniele, B;Bencivenga, A;Tinessa, V
通讯作者：
Tinessa, V
RASSF1A基因的启动子和第一个外显子在肝癌发生中的差异甲基化。
DOI：
10.1111/hepr.12449
发表时间：
2015-11
期刊：
Hepatology research : the official journal of the Japan Society of Hepatology
影响因子：
--
作者：
Jain S;Xie L;Boldbaatar B;Lin SY;Hamilton JP;Meltzer SJ;Chen SH;Hu CT;Block TM;Song W;Su YH
通讯作者：
Su YH
影响因子：
5.8
作者：
Sanyal, Arun J.;Yoon, Seung Kew;Lencioni, Riccardo
通讯作者：
Lencioni, Riccardo