Deep learning radiomics based on contrast enhanced computed tomography predicts microvascular invasion and survival outcome in early stage hepatocellular carcinoma

医学 无线电技术 队列 肝细胞癌 比例危险模型 接收机工作特性 放射科 阶段(地层学) 放射性武器 危险系数 肿瘤科 人工智能 内科学 置信区间 计算机科学 古生物学 生物
作者
Yuhan Yang,Yin Zhou,Chen Zhou,Xuelei Ma
出处
期刊:Ejso [Elsevier]
卷期号:48 (5): 1068-1077 被引量:23
标识
DOI:10.1016/j.ejso.2021.11.120
摘要

To evaluate the performance of a deep learning (DL)-based radiomics strategy on contrast-enhanced computed tomography (CT) to predict microvascular invasion (MVI) status and clinical outcomes, recurrence-free survival (RFS) and overall survival (OS) in patients with early stage hepatocellular carcinoma (HCC) receiving surgical resection.All 283 eligible patients were included retrospectively between January 2008 and December 2015, and assigned into the training cohort (n = 198) and the testing cohort (n = 85). We extracted radiomics features via handcrafted radiomics analysis manually and DL analysis of pretrained convolutional neural networks via transfer learning automatically. Support vector machine was adopted as the classifier. A clinical-radiological model for MVI status integrated significant clinical features and the radiological signature generated from the radiological model with the optimal area under the receiver operating characteristics curve (AUC) in the testing cohort. Otherwise, DL-based prognostic models were constructed in prediction of recurrence and mortality via Cox proportional hazard analysis.The clinical-radiological model for MVI represented an AUC of 0.909, accuracy of 96.47%, sensitivity of 90.91%, specificity of 97.30%, positive predictive value of 83.33%, and negative predictive value of 98.63% in the testing cohort. The clinical-radiological models for identification of RFS and OS outperformed prediction performance of the clinical model or the DL signature alone. The DL-based integrated model for prognostication showed great predictive value with significant classification and discrimination abilities after validation.The integrated DL-based radiomics models achieved accurate preoperative prediction of MVI status, and might facilitate predicting tumor recurrence and mortality in order to optimize clinical decisions for patients with early stage HCC.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
大幅提高文件上传限制,最高150M (2024-4-1)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
Yu发布了新的文献求助10
1秒前
ured发布了新的文献求助10
2秒前
Naveed完成签到,获得积分10
9秒前
Liway完成签到,获得积分20
9秒前
楠易完成签到,获得积分20
10秒前
10秒前
汉堡包应助骁悉采纳,获得10
11秒前
Bonnienuit发布了新的文献求助200
11秒前
共享精神应助东方神齐采纳,获得10
11秒前
12秒前
12秒前
Cat应助www采纳,获得10
13秒前
嘻嘻发布了新的文献求助10
15秒前
田様应助123采纳,获得10
17秒前
星辰大海应助务实安波采纳,获得10
17秒前
逍遥解牛发布了新的文献求助10
17秒前
shuiwuming完成签到 ,获得积分10
18秒前
沫沫完成签到 ,获得积分10
19秒前
LL发布了新的文献求助10
22秒前
楠易关注了科研通微信公众号
22秒前
呆萌语柳完成签到,获得积分10
22秒前
深情安青应助东方神齐采纳,获得10
24秒前
30秒前
30秒前
TRTHHRTZ应助DXY采纳,获得20
31秒前
嘟嘟发布了新的文献求助20
35秒前
35秒前
37秒前
李健的小迷弟应助lcjzbk采纳,获得30
38秒前
小爽发布了新的文献求助10
38秒前
丘比特应助东方神齐采纳,获得10
39秒前
39秒前
40秒前
勤恳惮完成签到,获得积分10
43秒前
43秒前
44秒前
薰硝壤应助心灵美的冷玉采纳,获得30
45秒前
45秒前
隐形曼青应助QXQ采纳,获得10
45秒前
nora完成签到 ,获得积分20
46秒前
高分求助中
LNG地下式貯槽指針(JGA指-107) 1000
LNG地上式貯槽指針 (JGA指 ; 108) 1000
QMS18Ed2 | process management. 2nd ed 600
LNG as a marine fuel—Safety and Operational Guidelines - Bunkering 560
How Stories Change Us A Developmental Science of Stories from Fiction and Real Life 500
九经直音韵母研究 500
Full waveform acoustic data processing 500
热门求助领域 (近24小时)
化学 医学 材料科学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 基因 遗传学 物理化学 催化作用 免疫学 细胞生物学 电极
热门帖子
关注 科研通微信公众号,转发送积分 2935440
求助须知:如何正确求助?哪些是违规求助? 2591040
关于积分的说明 6980414
捐赠科研通 2235974
什么是DOI,文献DOI怎么找? 1187421
版权声明 589879
科研通“疑难数据库(出版商)”最低求助积分说明 581288