医学成像中的诊断图像质量评估与分类：机遇与挑战
DIAGNOSTIC IMAGE QUALITY ASSESSMENT AND CLASSIFICATION IN MEDICAL IMAGING: OPPORTUNITIES AND CHALLENGES.
作者信息
Ma Jeffrey J, Nakarmi Ukash, Kin Cedric Yue Sik, Sandino Christopher M, Cheng Joseph Y, Syed Ali B, Wei Peter, Pauly John M, Vasanawala Shreyas S
机构信息
Department of Computing and Mathematical Sciences, California Institute of Technology.
Department of Radiology, Stanford University.
出版信息
Proc IEEE Int Symp Biomed Imaging. 2020 Apr;2020:337-340. doi: 10.1109/isbi45749.2020.9098735. Epub 2020 May 22.
Abstract
Magnetic Resonance Imaging (MRI) suffers from several artifacts, the most common of which are motion artifacts. These artifacts often yield images that are of non-diagnostic quality. To detect such artifacts, images are prospectively evaluated by experts for their diagnostic quality, which necessitates patient-revisits and rescans whenever non-diagnostic quality scans are encountered. This motivates the need to develop an automated framework capable of accessing medical image quality and detecting diagnostic and non-diagnostic images. In this paper, we explore several convolutional neural network-based frameworks for medical image quality assessment and investigate several challenges therein.
摘要
磁共振成像（MRI）存在多种伪影，其中最常见的是运动伪影。这些伪影常常产生质量无法用于诊断的图像。为了检测此类伪影，专家会对图像的诊断质量进行前瞻性评估，这就使得每当遇到质量无法用于诊断的扫描时，都需要患者再次就诊并重新扫描。这促使人们需要开发一个能够评估医学图像质量并检测诊断性和非诊断性图像的自动化框架。在本文中，我们探索了几种基于卷积神经网络的医学图像质量评估框架，并研究了其中的若干挑战。
相似文献
1
DIAGNOSTIC IMAGE QUALITY ASSESSMENT AND CLASSIFICATION IN MEDICAL IMAGING: OPPORTUNITIES AND CHALLENGES.医学成像中的诊断图像质量评估与分类：机遇与挑战
Proc IEEE Int Symp Biomed Imaging. 2020 Apr;2020:337-340. doi: 10.1109/isbi45749.2020.9098735. Epub 2020 May 22.
2
Automatic MR image quality evaluation using a Deep CNN: A reference-free method to rate motion artifacts in neuroimaging.使用深度卷积神经网络的自动磁共振图像质量评估：一种用于评定神经影像学中运动伪影的无参考方法。
Comput Med Imaging Graph. 2021 Jun;90:101897. doi: 10.1016/j.compmedimag.2021.101897. Epub 2021 Mar 11.
3
Automatic brain MRI motion artifact detection based on end-to-end deep learning is similarly effective as traditional machine learning trained on image quality metrics.基于端到端深度学习的自动脑 MRI 运动伪影检测与基于图像质量指标的传统机器学习训练同样有效。
Med Image Anal. 2023 Aug;88:102850. doi: 10.1016/j.media.2023.102850. Epub 2023 May 23.
4
Reduction of respiratory motion artifacts in gadoxetate-enhanced MR with a deep learning-based filter using convolutional neural network.基于卷积神经网络的深度学习滤波器在钆塞酸增强磁共振中减少呼吸运动伪影。
Eur Radiol. 2020 Nov;30(11):5923-5932. doi: 10.1007/s00330-020-07006-1. Epub 2020 Jun 17.
5
Mitigation of motion-induced artifacts in cone beam computed tomography using deep convolutional neural networks.使用深度卷积神经网络减轻锥束计算机断层扫描中运动引起的伪影。
Med Phys. 2023 Oct;50(10):6228-6242. doi: 10.1002/mp.16405. Epub 2023 Apr 11.
6
Automatic detection of Gibbs artefact in MR images with transfer learning approach.基于迁移学习方法的磁共振图像中吉布斯伪影自动检测
Technol Health Care. 2023;31(1):239-246. doi: 10.3233/THC-220234.
7
Retrospective correction of motion-affected MR images using deep learning frameworks.使用深度学习框架对受运动影响的磁共振图像进行回顾性校正。
Magn Reson Med. 2019 Oct;82(4):1527-1540. doi: 10.1002/mrm.27783. Epub 2019 May 13.
8
Application of deep learning-based image reconstruction in MR imaging of the shoulder joint to improve image quality and reduce scan time.基于深度学习的图像重建在肩关节磁共振成像中的应用，以提高图像质量并减少扫描时间。
Eur Radiol. 2023 Mar;33(3):1513-1525. doi: 10.1007/s00330-022-09151-1. Epub 2022 Sep 27.
9
Brain MRI artefact detection and correction using convolutional neural networks.使用卷积神经网络进行脑 MRI 伪影检测和校正。
Comput Methods Programs Biomed. 2021 Feb;199:105909. doi: 10.1016/j.cmpb.2020.105909. Epub 2020 Dec 23.
10
A convolutional neural network to filter artifacts in spectroscopic MRI.一种用于过滤磁共振波谱成像伪影的卷积神经网络。
Magn Reson Med. 2018 Nov;80(5):1765-1775. doi: 10.1002/mrm.27166. Epub 2018 Mar 9.
引用本文的文献
1
Automated quality evaluation of dental panoramic radiographs using deep learning.使用深度学习对牙科全景X光片进行自动质量评估。
Imaging Sci Dent. 2025 Jun;55(2):175-188. doi: 10.5624/isd.20240232. Epub 2025 Apr 10.
2
Applications of Artificial Intelligence for Pediatric Cancer Imaging.人工智能在儿科癌症成像中的应用。
AJR Am J Roentgenol. 2024 Aug;223(2):e2431076. doi: 10.2214/AJR.24.31076. Epub 2024 May 23.
3
Quantifying MR head motion in the Rhineland Study - A robust method for population cohorts.量化 Rhineland 研究中的磁共振头部运动 - 一种针对人群队列的稳健方法。
Neuroimage. 2023 Jul 15;275:120176. doi: 10.1016/j.neuroimage.2023.120176. Epub 2023 May 18.
4
Pre-training via Transfer Learning and Pretext Learning a Convolutional Neural Network for Automated Assessments of Clinical PET Image Quality.通过迁移学习和自监督学习预训练卷积神经网络用于临床PET图像质量的自动评估
IEEE Trans Radiat Plasma Med Sci. 2023 Apr;7(4):372-381. doi: 10.1109/TRPMS.2022.3231702.
5
Robust Classification and Detection of Big Medical Data Using Advanced Parallel -Means Clustering, YOLOv4, and Logistic Regression.使用先进的并行均值聚类、YOLOv4和逻辑回归对大型医学数据进行稳健分类与检测
Life (Basel). 2023 Mar 3;13(3):691. doi: 10.3390/life13030691.
6
An artificial intelligence-driven image quality assessment system for whole-body [F]FDG PET/CT.基于人工智能的全身 [F]FDG PET/CT 图像质量评估系统。
Eur J Nucl Med Mol Imaging. 2023 Apr;50(5):1318-1328. doi: 10.1007/s00259-022-06078-z. Epub 2022 Dec 19.
7
Automatic Artifact Detection Algorithm in Fetal MRI.胎儿磁共振成像中的自动伪影检测算法
Front Artif Intell. 2022 Jun 16;5:861791. doi: 10.3389/frai.2022.861791. eCollection 2022.
8
A Brief Survey on No-Reference Image Quality Assessment Methods for Magnetic Resonance Images.磁共振图像无参考图像质量评估方法简述
J Imaging. 2022 Jun 4;8(6):160. doi: 10.3390/jimaging8060160.
9
Evaluation of motion artifacts in brain magnetic resonance images using convolutional neural network-based prediction of full-reference image quality assessment metrics.基于卷积神经网络的全参考图像质量评估指标预测对脑磁共振图像中运动伪影的评估
J Med Imaging (Bellingham). 2022 Jan;9(1):015502. doi: 10.1117/1.JMI.9.1.015502. Epub 2022 Jan 21.
10
Five-star rating system for acceptable quality and dose in CT.CT 可接受质量和剂量的五星评级系统。
Eur Radiol. 2021 Dec;31(12):9161-9163. doi: 10.1007/s00330-021-08112-4. Epub 2021 Jun 11.
本文引用的文献
1
Deep Generative Adversarial Neural Networks for Compressive Sensing MRI.用于压缩感知 MRI 的深度生成对抗神经网络。
IEEE Trans Med Imaging. 2019 Jan;38(1):167-179. doi: 10.1109/TMI.2018.2858752. Epub 2018 Jul 23.
2
NIMA: Neural Image Assessment.NIMA：神经影像评估。
IEEE Trans Image Process. 2018 Apr 30. doi: 10.1109/TIP.2018.2831899.
3
Automated reference-free detection of motion artifacts in magnetic resonance images.磁共振图像中运动伪影的自动无参考检测
MAGMA. 2018 Apr;31(2):243-256. doi: 10.1007/s10334-017-0650-z. Epub 2017 Sep 20.
4
Automated image quality evaluation of T -weighted liver MRI utilizing deep learning architecture.利用深度学习架构进行 T 加权肝脏 MRI 的自动图像质量评估。
J Magn Reson Imaging. 2018 Mar;47(3):723-728. doi: 10.1002/jmri.25779. Epub 2017 Jun 3.
5
Artifacts in magnetic resonance imaging.磁共振成像中的伪影。
Pol J Radiol. 2015 Feb 23;80:93-106. doi: 10.12659/PJR.892628. eCollection 2015.
6
Motion artifacts in MRI: A complex problem with many partial solutions.磁共振成像中的运动伪影：一个复杂问题，有许多部分解决方案。
J Magn Reson Imaging. 2015 Oct;42(4):887-901. doi: 10.1002/jmri.24850. Epub 2015 Jan 28.
7
Reproduction of motion artifacts for performance analysis of prospective motion correction in MRI.运动伪影的再现用于前瞻性运动校正 MRI 中的性能分析。
Magn Reson Med. 2014 Jan;71(1):182-90. doi: 10.1002/mrm.24645. Epub 2013 Feb 25.
8
Navigated abdominal T1-W MRI permits free-breathing image acquisition with less motion artifact.导航式腹部 T1-W MRI 可实现自由呼吸下的图像采集，运动伪影更少。
Pediatr Radiol. 2010 Mar;40(3):340-4. doi: 10.1007/s00247-009-1502-4.
9
No-reference image quality metrics for structural MRI.用于结构磁共振成像的无参考图像质量指标。
Neuroinformatics. 2006 Summer;4(3):243-62. doi: 10.1385/NI:4:3:243.
Zotero 插件