Performance comparison between MRI and CT for local staging of sigmoid and descending colon cancer

医学 降结肠 乙状窦函数 放射科 结直肠癌 乙状结肠 结肠疾病 磁共振成像 核医学 癌症 内科学 人工智能 直肠 计算机科学 人工神经网络
作者
Li-Heng Liu,Han Lv,Zhenchang Wang,Sheng-Xiang Rao,Mengsu Zeng
出处
期刊:European Journal of Radiology [Elsevier BV]
卷期号:121: 108741-108741 被引量:22
标识
DOI:10.1016/j.ejrad.2019.108741
摘要

Purpose To compare the diagnostic performance of MRI and CT for local staging of sigmoid and descending colon cancer, with pathological results as the reference standard. Method This retrospective study included 116 patients with sigmoid or descending colon cancer who underwent both MRI and CT before surgery. MRI and CT images were separately reviewed by two independent and blinded radiologists to assess the following features: T-stage, presence of extramural extension (T3-4 disease), lymph node metastases (N+), and extramural vascular invasion (EMVI+). Diagnostic performance with sensitivity and specificity for detecting positive status (T3-4, N+ or EMVI+) were assessed using receiver-operating-characteristic (ROC) curve, and compared between MRI and CT. Results MRI achieved correct T-stage in 81 of 116 patients (69.8 %) while CT in 66 (56.9 %). For detecting T3-4 disease, MRI showed better performance than CT with area under the curve (AUC) of 0.888 versus 0.712 (P = 0.002) and specificity of 81.82 % versus 54.6 % (P = 0.011). No significance was found in sensitivity between two modalities (89.2 % versus 83.1 %, P = 0.302). For detecting N+ disease, performance of MRI and CT were similar (AUC, 0.670 versus 0.650, P = 0.412). For detecting EMVI+, MRI showed better performance than CT (AUC, 0.780 versus 0.575, P = 0.012) with significantly higher sensitivity (68.6 % versus 40.0 %, P = 0.031) and similar specificity (both are 84.3 %). Conclusions MRI may offer more superior diagnostic performance than CT for detecting T3-4 disease and EMVI, thereby supporting its alternative application to CT in local staging of colon cancer.
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
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
王晨昕发布了新的文献求助30
刚刚
1秒前
苏苏完成签到,获得积分10
1秒前
2秒前
思策完成签到,获得积分10
2秒前
dyc发布了新的文献求助10
2秒前
徐小锤发布了新的文献求助10
2秒前
dashuaibiu完成签到,获得积分10
2秒前
boshazhiwu完成签到 ,获得积分10
3秒前
3秒前
脑洞疼应助Ly啦啦啦采纳,获得10
3秒前
3秒前
3秒前
4秒前
星辰大海应助搁浅采纳,获得10
4秒前
frenchfriespie完成签到,获得积分10
4秒前
4秒前
4秒前
5秒前
希望天下0贩的0应助新明采纳,获得10
5秒前
安静青亦完成签到,获得积分10
6秒前
6秒前
YY完成签到 ,获得积分10
6秒前
6秒前
李爱国应助思策采纳,获得10
6秒前
7秒前
科研通AI5应助lzj采纳,获得10
7秒前
ABC2023完成签到,获得积分10
8秒前
pxin发布了新的文献求助10
8秒前
苗条的凝梦完成签到,获得积分10
8秒前
魔幻的山柏完成签到,获得积分10
8秒前
什么东西完成签到,获得积分10
9秒前
一帆风顺发布了新的文献求助10
9秒前
9秒前
聪明诗槐完成签到,获得积分10
9秒前
weddcf发布了新的文献求助10
9秒前
早日暴富完成签到,获得积分20
9秒前
oboy应助泽灵采纳,获得10
9秒前
觞f完成签到,获得积分10
10秒前
Hello应助知性的奎采纳,获得10
10秒前
高分求助中
Thinking Small and Large 500
Algorithmic Mathematics in Machine Learning 500
Getting Published in SSCI Journals: 200+ Questions and Answers for Absolute Beginners 300
On translated images, stereotypes and disciplines 200
Deciphering Earth's History: the Practice of Stratigraphy 200
New Syntheses with Carbon Monoxide 200
Quanterion Automated Databook NPRD-2023 200
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 物理 生物化学 纳米技术 计算机科学 化学工程 内科学 复合材料 物理化学 电极 遗传学 量子力学 基因 冶金 催化作用
热门帖子
关注 科研通微信公众号,转发送积分 3834256
求助须知:如何正确求助?哪些是违规求助? 3376847
关于积分的说明 10495379
捐赠科研通 3096271
什么是DOI,文献DOI怎么找? 1704904
邀请新用户注册赠送积分活动 820296
科研通“疑难数据库(出版商)”最低求助积分说明 771940