Deep Learning-Based Automatic Detection of Brain Metastases in Heterogenous Multi-Institutional Magnetic Resonance Imaging Sets: An Exploratory Analysis of NRG-CC001

卷积神经网络 Sørensen–骰子系数 人工智能 规范化(社会学) 模式识别(心理学) 掷骰子 试验装置 磁共振成像 分割 医学 深度学习 相似性(几何) 数据集 交叉验证 计算机科学 图像分割 统计 放射科 数学 图像(数学) 社会学 人类学
作者
Ying Liang,Karen Lee,Joseph Bovi,Joshua D. Palmer,Paul D. Brown,Vinai Gondi,Wolfgang A. Tomé,Tammie L.S. Benzinger,Minesh P. Mehta,X Allen Li
出处
期刊:International Journal of Radiation Oncology Biology Physics [Elsevier BV]
卷期号:114 (3): 529-536 被引量:5
标识
DOI:10.1016/j.ijrobp.2022.06.081
摘要

Deep learning-based algorithms have been shown to be able to automatically detect and segment brain metastases (BMs) in magnetic resonance imaging, mostly based on single-institutional data sets. This work aimed to investigate the use of deep convolutional neural networks (DCNN) for BM detection and segmentation on a highly heterogeneous multi-institutional data set.A total of 407 patients from 98 institutions were randomly split into 326 patients from 78 institutions for training/validation and 81 patients from 20 institutions for unbiased testing. The data set contained T1-weighted gadolinium and T2-weighted fluid-attenuated inversion recovery magnetic resonance imaging acquired on diverse scanners using different pulse sequences and various acquisition parameters. Several variants of 3-dimensional U-Net based DCNN models were trained and tuned using 5-fold cross validation on the training set. Performances of different models were compared based on Dice similarity coefficient for segmentation and sensitivity and false positive rate (FPR) for detection. The best performing model was evaluated on the test set.A DCNN with an input size of 64 × 64 × 64 and an equal number of 128 kernels for all convolutional layers using instance normalization was identified as the best performing model (Dice similarity coefficient 0.73, sensitivity 0.86, and FPR 1.9) in the 5-fold cross validation experiments. The best performing model demonstrated consistent behavior on the test set (Dice similarity coefficient 0.73, sensitivity 0.91, and FPR 1.7) and successfully detected 7 BMs (out of 327) that were missed during manual delineation. For large BMs with diameters greater than 12 mm, the sensitivity and FPR improved to 0.98 and 0.3, respectively.The DCNN model developed can automatically detect and segment brain metastases with reasonable accuracy, high sensitivity, and low FPR on a multi-institutional data set with nonprespecified and highly variable magnetic resonance imaging sequences. For large BMs, the model achieved clinically relevant results. The model is robust and may be potentially used in real-world situations.

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
大个的应助被风趣的小鸽子采纳,获得10
1秒前
十六完成签到,获得积分10
1秒前
1秒前
机灵的明杰完成签到 ,获得积分10
1秒前
Ava的应助被麤七采纳,获得10
1秒前
xx完成签到,获得积分10
1秒前
2秒前
ZJB的应助被火星上的蜻蜓采纳,获得10
2秒前
2秒前
无极微光的应助被青橙采纳,获得20
3秒前
3秒前
3秒前
3秒前
70000完成签到 ,获得积分10
4秒前
Samuel发布了新的文献求助10
4秒前
CuO完成签到,获得积分10
4秒前
4秒前
wbj发布了新的文献求助10
4秒前
yu发布了新的文献求助20
4秒前
adasda发布了新的文献求助10
4秒前
结实的纹完成签到,获得积分10
5秒前
大大大大大阳完成签到,获得积分10
5秒前
5秒前
5秒前
风趣的元槐完成签到,获得积分10
5秒前
an发布了新的文献求助30
6秒前
艾乐西沟完成签到,获得积分10
6秒前
南檬发布了新的文献求助10
6秒前
guoguo发布了新的文献求助10
6秒前
实验昏了头完成签到,获得积分10
6秒前
李健的粉丝团团长的应助被pb采纳,获得10
7秒前
YuJue的应助被yao采纳,获得20
7秒前
Lucas的应助被adasda采纳,获得10
7秒前
风趣青曼完成签到,获得积分10
7秒前
7秒前
隐形曼青的应助被Hj采纳,获得10
8秒前
8秒前
8秒前
8秒前
9秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1500
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
八旗制度の研究 600
Analyses de reconnaissances de dettes passées devant les échevins d'Ypres (1249-1291) 600
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8159929
求助须知:如何正确求助?哪些是违规求助? 9555266
关于积分的说明 21726335
捐赠科研通 7749807
什么是DOI,文献DOI怎么找? 3427310
关于科研通互助平台的介绍 2552391
邀请新用户注册赠送积分活动 2471654