Received:2019-09-25Online:2019-12-20 作者简介 About authors 俞益洲,1970年生,深睿医疗首席科学家,曾任美国伊利诺依大学香槟分校终身教授。加州大学伯克利分校计算机博士,美国计算机协会杰出科学家,IEEE Fellow。已在具有影响力的国际会议和期刊发表学术论文150余篇,多次获最佳论文奖。担任图像、视觉计算领域重要国际会议程序委员会委员或主席,并在多个国际学术期刊担任副主编。 本文承担工作为:AI技术及其在医学影像中的应用分析与讨论以及全文统筹。 Yu Yizhou, born in 1970, is a full professor at the University of Hong Kong. He was a tenured professor at University of Illinois, Urbana-Champaign (UIUC). He received his PhD degree in computer science at University of California, Berkeley. He is an IEEE fellow and ACM distinguished member. He has published more than 150 academic papers in influential international conferences and journals and has won many best paper awards. He serves as a member or chairman at important international conferences in the field of image processing and computer vision and serves as deputy editor of several international academic journals. Role in this paper: conceptualized and organized the review on the application of deep learning in medical imaging analysis. E-mail: yizhouy@acm.org
马杰超,1991年生,深睿医疗AI研究院机器学习研究员,中山大学计算机硕士。主要研究方向为AI技术在医学图像上的应用。 本文承担工作为:调研、分析与讨论DL在脑卒中、肺结节、肺栓塞、乳腺疾病中的应用研究。 Ma Jiechao, born in 1991, is a machine learning researcher of Deepwise AI Lab and a Master of Computer Science from Sun Yat-sen University. His main research direction is the application of artificial intelligence technology in medical image. Role in this paper: reviewed and analyzed the research applications of deep learning in stroke, pulmonary nodules, pulmonary embolism and breast diseases. E-mail:majch7@mail2.sysu.edu.cn
石德君,1990年生,深睿医疗AI研究院机器学习研究员,北京大学医学部理学学士。主要研究方向为DL和医学影像分析。 本文承担工作为:调研、分析与讨论DL在糖网、肺结核、骨龄估计中的应用研究。 Shi Dejun, born in 1990, he graduated from Peking University Health Science Center, and is a machine learning researcher at Deepwise AI Lab. His research interests are deep learning and medical imaging analysis. Role in this paper: reviewed and analyzed the research applications of deep learning in diabetic retinopathy, pulmonary tuberculosis, and bone age assessment. E-mail:shidejun@deepwise.com
周振,1990年生,中国科学院自动化研究所博士,深睿医疗高级算法研究员。主要研究方向为DL和医学影像分析。 本文承担工作为:制定论文结构,撰写前言、DL简介和结论与展望。 Zhou Zhen, born in 1990, holds a PhD degree from Institute of Automation, Chinese Academy of Sciences. His research focuses on deep learning and medical imaging analysis. Role in this paper: outlined the review and drafted the sections including overview, introduction to deep learning and conclusion. E-mail:zhouzhen@deepwise.com
Abstract [Objective] This paper reviews the recent progress of deep learning researches and applications in medical image analysis. [Coverage] Relevant papers were first retrieved by keyword search and then by citation screening. [Methods] Deep learning based on convolutional neural networks is briefly introduced. Then, we review the diagnostic performance of deep learning on medical images in recent years with respect to different types of diseases, such as stroke, pulmonary nodules and bone age estimation. [Results] Deep learning for medical image interpretation has demonstrated advantages in many aspects, including accuracy, speed, stability and scalability. Meanwhile, existing problems may hinder clinical adoption of deep learning, such as dependence on a large amount of labelled data, inconsistent labeling standards, poor generalizability and interpretability of deep learning methods. [Limitations] There may be omissions of the retrieved literature, and it is impossible to compare the performance of the same deep learning model across different diseases. [Conclusions] Powerful artificial intelligence can improve the efficiency and accuracy of image interpretations for radiologists, but artificial intelligence is not perfect. Before being widely adopted in medical image interpretation, deep learning methods need more verification in real applications. Keywords:deep learning;medical imaging;survey
PDF (13574KB)元数据多维度评价相关文章导出EndNote|Ris|Bibtex收藏本文 本文引用格式 俞益洲, 马杰超, 石德君, 周振. 深度学习在医学影像分析中的应用综述. 数据与计算发展前沿[J], 2019, 1(2): 37-52 doi:10.11871/jfdc.issn.2096-742X.2019.02.004 Yu Yizhou. Application of Deep Learning in Medical Imaging Analysis: A Survey. Frontiers of Data and Computing[J], 2019, 1(2): 37-52 doi:10.11871/jfdc.issn.2096-742X.2019.02.004
近年来,多位国内外****应用DL技术探索新的骨龄评估自动化方案,表明了DL技术在骨龄评估方面的巨大潜力。2017年RSNA举行了迄今最大规模的AI骨龄评估挑战赛(RSNA 2017 Bone Age Challenge)[82]。冠军团队Mark Cicoro和Alexander Bilbily将年龄预测处理成单纯的数值回归问题,预测结果的平均偏差降低到4.265个月。
Table 1 表1 表1在多种医学影像分析中DL与传统方法和人类的表现对比 Table 1The performance comparison of DL models with traditional methods and humans on multiple applications in medical imaging
PapademetrisXDeLorenzoC, FlossmannS , et al. From medical image computing to computer‐aided intervention: development of a research interface for image-guided navigation [J]. The International Journal of Medical Robotics and Computer Assisted Surgery, 2009,5(2):147-157. [本文引用: 1]
LakhaniP, SundaramB . Deep Learning at Chest Radiography: Automated Classification of Pulmonary Tuberculosis by Using Convolutional Neural Networks .[J]. Radiology, 2017: 574-582. [本文引用: 3]
ZhuW., LiuC., FanW ., et al. DeepLung: Deep 3D Dual Path Nets for Automated Pulmonary Nodule Detection and Classification IEEE WACV, 2018. [本文引用: 2]
GeertLitjens, ThijsKooi , et al. A survey on deep learning in medical image analysis [J]. Medical Image Analysis, 2017: 60-88. [本文引用: 2]
DalalNavneet, and BillTriggs . Histograms of oriented gradients for human detection [C]. Computer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer Society Conference on. Vol. 1. IEEE, 2005. [本文引用: 1]
Lowe DavidG . Distinctive image features from scale-invariant keypoints [J].International journal of computer vision 60. 2(2004):91-110. [本文引用: 1]
SmistadE, Falch TL, BozorgiM , et al. Medical image segmentation on GPUs - A comprehensive review [J]. Medical Image Analysis, 2015,20(1):1-18. [本文引用: 1]
EstevaAndre , et al. Dermatologist-level classification of skin cancer with deep neural networks [J].Nature 542. 7639(2017):115. [本文引用: 2]
RenS, HeK, GirshickR, et al. Faster r-cnn: Towards real-time object detection with region proposal networks [C]// Advances in neural information processing systems, 2015: 91-99. [本文引用: 1]
LiuW, AnguelovD, ErhanD, et al. Ssd: Single shot multibox detector [C]// European conference on computer vision. Springer, Cham, 2016: 21-37. [本文引用: 1]
RedmonJ, DivvalaS, GirshickR, et al. You only look once: Unified, real-time object detection [C]// Proceedings of the IEEE conference on computer vision and pattern recognition, 2016: 779-788. [本文引用: 1]
KongB, ZhanY, Shin MC, et al. Recognizing End-Diastole and End-Systole Frames via Deep Temporal Regression Network [C]. medical image computing and computer assisted intervention, 2016: 264-272. [本文引用: 1]
SundermeyerM, SchlüterR, NeyH . LSTM neural networks for language modeling [C]// Thirteenth annual conference of the international speech communication association, 2012. [本文引用: 1]
YanK, BagheriM, Summers RM . 3d context enhanced region-based convolutional neural network for end-to-end lesion detection [C]// International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2018: 511-519. [本文引用: 1]
RonnebergerO, FischerP, BroxT . U-net: Convolutional networks for biomedical image segmentation [C]// International Conference on Medical image computing and computer-assisted intervention. Springer, Cham, 2015: 234-241. [本文引用: 1]
MilletariF, NavabN, Ahmadi SA . V-net: Fully convolutional neural networks for volumetric medical image segmentation [C]// 2016 Fourth International Conference on 3D Vision (3DV). IEEE, 2016: 565-571. [本文引用: 1]
?i?ek?, AbdulkadirA, Lienkamp SS, et al. 3D U-Net: learning dense volumetric segmentation from sparse annotation [C]// International conference on medical image computing and computer-assisted intervention. Springer, Cham, 2016: 424-432. [本文引用: 1]
RueckertD, Schnabel JA . Medical image registration [M] //Biomedical image processing. Springer, Berlin, Heidelberg, 2010: 131-154. [本文引用: 1]
SimonovskyM, GutierrezbeckerB, MateusD, et al. A Deep Metric for Multimodal Registration [C]. medical image computing and computer assisted intervention, 2016: 10-18. [本文引用: 1]
MiaoS, Wang ZJ, LiaoR , et al. A CNN Regression Approach for Real-Time 2D/3D Registration [J]. IEEE Transactions on Medical Imaging, 2016,35(5):1352-1363. [本文引用: 1]
JaderbergM, SimonyanK, ZissermanA, et al. Spatial transformer networks [C]. neural information processing systems, 2015: 2017-2025. [本文引用: 1]
HaskinsG, KrugerU, YanP . Deep Learning in Medical Image Registration: A Survey [J]. arXiv:1903.02026. 2019. [本文引用: 1]
KrupinskiJ, KaluzaJ, KumarP , et al. Role of angiogenesis in patients with cerebral ischemic stroke [J]. Stroke, 1994,25(9):1794-1798. [本文引用: 1]
HuXDeSilva T M, ChenJ , et al. Cerebral Vascular Disease and Neurovascular Injury in Ischemic Stroke [J]. Circulation Research, 2017,120(3):449-471. [本文引用: 1]
ChilamkurthyS, GhoshR, TanamalaS , et al. Development and Validation of Deep Learning Algorithms for Detection of Critical Findings in Head CT Scans [J]. arXiv preprint arXiv:1803.05854, 2018. [本文引用: 1]
Chang PD, KuoyE, GrinbandJ , et al. Hybrid 3D/2D convolutional neural network for hemorrhage evaluation on head CT [J]. American Journal of Neuroradiology, 2018,39(9):1609-1616. [本文引用: 1]
ZhangR, ZhaoL, LouW , et al. Automatic Segmentation of Acute Ischemic Stroke from DWI using 3D Fully Convolutional DenseNets [J]. IEEE Transactions on Medical Imaging, 2018: 1-1. [本文引用: 2]
LiuQ, ChengX, ZhangQ, ZhouC, LiX, WangS, KongmingLiang, GuangmingLu , Automatic Segmentation for Acute Ischemic Stroke from DWI Using Deep Convolutional Neural Networks [J]. European Society of Radiology (2019). [本文引用: 1]
MocciaS, Momi ED, Hadji SE , et al. Blood vessel segmentation algorithms — Review of methods, datasets and evaluation metrics [J]. Computer Methods & Programs in Biomedicine, 2018,158:71-91. [本文引用: 1]
TettehG, EfremovV, Forkert ND , et al. DeepVesselNet: Vessel Segmentation, Centerline Prediction, and Bifurcation Detection in 3-D Angiographic Volumes [J]. 2018. [本文引用: 1]
Fong DS, Aiello LP, Gardner TW , et al. Retinopathy in Diabetes [J]. Diabetes Care, 2004,27(1). [本文引用: 1]
JonesS, Edwards RT . Diabetic retinopathy screening: a systematic review of the economic evidence [J]. Diabet Med, 2010,27(3):249-256. [本文引用: 1]
Nielsen KB, Lautrup MLAndersenJ K H , et al. Deep Learning-based Algorithms in Screening of Diabetic Retinopathy: A Systematic Review of Diagnostic Performance .[J]. Ophthalmology Retina, 2018. [本文引用: 1]
QuellecG, CharriereK, BoudiY , et al. Deep image mining for diabetic retinopathy screening [J]. Medical Image Analysis, 2017,39:178-193. [本文引用: 2]
TakahashiH, TampoH, AraiY , et al. Applying artificial intelligence to disease staging: deep learning for improved staging of diabetic retinopathy [J]. PLoS ONE, 2017,12:e0179790. [本文引用: 1]
JiangZ., ZhangH., WangY ., et al. Retinal blood vessel segmentation using fully convolutional network with transfer learning [J]. Computerized Medical Imaging and Graphics, 2018,68:1-15. [本文引用: 1]
Fraz MM, RemagninoP, HoppeA , et al. Blood vessel segmentation methodologies in retinal images-a survey [J]. Computer methods and programs in biomedicine, 2012,108(1):407-433. [本文引用: 2]
Xiao ZT, Zhang XP, GengL , et al. Automatic non-proliferative diabetic retinopathy screening system based on color fundus image [J]. BioMedical Engineering OnLine, 2017,16:122. [本文引用: 1]
Prenta?i?P., Lon?ari?S . Detection of exudates in fundus photographs using deep neural networks and anatomical landmark detection fusion [J]. Computer Methods and Programs in Biomedicine, 2016,137:281-292. [本文引用: 1]
Bray FI, FerlayJ, SoerjomataramI , et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries [J]. CA: A Cancer Journal for Clinicians, 2018,68(6):394-424. [本文引用: 1]
Siegel RL, Miller KD, AhmedinJemal . Cancer statistics, 2018 [J]. Ca A Cancer Journal for Clinicians, 2018,68(1):11. [本文引用: 1]
Van GinnekenB, ArmatoIII S G, deHoop B , et al. Comparing and combining algorithms for computer-aided detection of pulmonary nodules in computed tomography scans: the ANODE09 study [J]. Medical image analysis, 2010,14(6):707-722. [本文引用: 2]
Setio A AA, JacobsC, GelderblomJ , et al. Automatic detection of large pulmonary solid nodules in thoracic CT images [J]. Medical Physics, 2015,42. [本文引用: 1]
ZhuW, LiuC, FanW, et al. DeepLung: Deep 3D Dual Path Nets for Automated Pulmonary Nodule Detection and Classification [C]. IEEE Winter Conference on Applications of Computer Vision (WACV), 2018: 673-681. [本文引用: 2]
Setio A AA, TraversoA, De BelT , et al. Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the LUNA16 challenge [J]. Medical image analysis, 2017,42:1-13. [本文引用: 2]
WangS, ZhouM, GevaertO, et al. A multi-view deep convolutional neural networks for lung nodule segmentation [C]// 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE, 2017. [本文引用: 1]
NamC, KimJ, LeeK . Lung nodule segmentation with convolutional neural network trained by simple diameter information [C]// 1st Conference on Medical Imaging with Deep Learning, 2018. [本文引用: 2]
HaichaoC, fHongL, EnminS, Chih-ChengH, GuangzhiM , et al. Dual-branch residual network for lung nodule segmentation Arxiv 2019 [1905.080413]. [本文引用: 2]
ShenS., HanS., AberleD ., et al. An Interpretable Deep Hierarchical Semantic Convolutional Neural Network for Lung Nodule Malignancy Classification Arxiv 2018 [1806.00712]. [本文引用: 1]
WuB, ZhouZ, WangJ , et al. Joint Learning for Pulmonary Nodule Segmentation, Attributes and Malignancy Prediction [J]. 2018. [本文引用: 1]
MundherA, BoonL, WaiY , et al. Lung Nodule Classification using Deep Local-Global Networks Arxiv 2019 [1904.10126]. [本文引用: 1]
Blackmon KN, FlorinC, BogoniL , et al. Computer-aided detection of pulmonary embolism at CT pulmonary angiography: can it improve performance of inexperienced readers? [J]. European Radiology, 2011,21(6):1214-1223. [本文引用: 1]
RuccoM, SousarodriguesD, MerelliE , et al. Neural hypernetwork approach for pulmonary embolism diagnosis [J]. BMC Research Notes, 2015,8(1):617-617. [本文引用: 1]
BiJ, LiangJ . Multiple Instance Learning of Pulmonary Embolism Detection with Geodesic Distance along Vascular Structure [C]. computer vision and pattern recognition, 2007: 1-8. [本文引用: 1]
Chen MC, Ball RL, YangL , et al. Deep Learning to Classify Radiology Free-Text Reports [J]. Radiology, 2017. [本文引用: 1]
World HealthOrganization . Global tuberculosis report 2018 [R]. Geneva, Switzerland: World Health Organization, 2018. [本文引用: 1]
PandeT, CohenC, PaiM , et al. Computer Aided Detection of Pulmonary Tuberculosis on Digital Chest Radiographs: a systematic review [J]. The International Journal of Tuberculosis and Lung Disease, 2016,20(9):1226. [本文引用: 1]
AntaniS . Automated Detection of Lung Diseases in Chest X-Rays. A Report to the Board of Scientific Counselors [R]. US National Library of Medicine. https://lhncbc.nlm.nih.gov/system/files/pub9126.pdf. Published April 2015. Accessed September 20, 2016. URL [本文引用: 1]
JaegerS, KarargyrisA, CandemirS , et al. Automatic screening for tuberculosis in chest radiographs: a survey [J]. Quantitative imaging in medicine and surgery, 2013,3(2):89-99. [本文引用: 1]
HwangS, Kim HE, JeongJ, et al. A novel approach for tuberculosis screening based on deep convolutional neural networks [C]// Medical Imaging 2016: Computer-Aided Diagnosis. International Society for Optics and Photonics, 2016. [本文引用: 1]
JemalA, SiegelR, WardE , et al. Cancer statistics, 2008 [J]. Ca A Cancer Journal for Clinicians, 2008,58(2):71. [本文引用: 1]
Gram IT, FunkhouserE, TabárL . The Tabar classification of mammographic parenchymal patterns [J]. European Journal of Radiology, 1997,24(2):131-6. [本文引用: 1]
KooiT, LitjensG, Van GinnekenB , et al. Large scale deep learning for computer aided detection of mammographic lesions [J]. Medical image analysis, 2017,35:303-312. [本文引用: 1]
ArevaloJ, González FA, Ramos-PollánR , et al. Representation learning for mammography mass lesion classification with convolutional neural networks [J]. Computer methods and programs in biomedicine, 2016,127:248-257. [本文引用: 1]
Fotin SV, YinY, HaldankarH, et al. Detection of soft tissue densities from digital breast tomosynconfproc: comparison of conventional and deep learning approaches [C]// Medical Imaging 2016: Computer-Aided Diagnosis. International Society for Optics and Photonics, 2016,9785:97850X. [本文引用: 2]
B BuciuI , Gacsadi A. Directional features for automatic tumor classification of mammogram images [J]. Biomedical Signal Processing & Control, 2011,6(4):370-378. [本文引用: 1]
Shastri AA, TamrakarD, AhujaK . Density-wise two stage mammogram classification using texture exploiting descriptors [J]. Expert Systems with Applications, 2018,99:71-82. [本文引用: 1]
BriaA, KarssemeijerN, TortorellaF . Learning from unbalanced data: a cascade-based approach for detecting clustered microcalcifications [J]. Medical image analysis, 2014,18(2):241-252. [本文引用: 1]
CaiG, GuoY, ZhangY, et al. A fully automatic microcalcification detection approach based on deep convolution neural network [C]// Medical Imaging 2018: Computer-Aided Diagnosis. International Society for Optics and Photonics, 2018,10575:105752Q. [本文引用: 1]
ZhangF, LuoL, SunX, et al. Cascaded Generative and Discriminative Learning for Microcalcification Detection in Breast Mammograms [C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2019: 12578-12586. [本文引用: 1]
Iglovikov VI, RakhlinA, Kalinin AA , et al. Paediatric Bone Age Assessment Using Deep Convolutional Neural Networks [J]. arXiv: Computer Vision and Pattern Recognition, 2018: 300-308. [本文引用: 1]
Son SJ, SongY, KimN , et al. TW3-Based Fully Automated Bone Age Assessment System Using Deep Neural Networks [J]. IEEE Access, 2019: 33346-33358. [本文引用: 1]
MayeuxR, SternY . Epidemiology of Alzheimer Disease [J]. Cold Spring Harbor Perspectives in Medicine, 2012,2(8):1-19. [本文引用: 1]
DurstewitzD, KoppeG, Meyer-LindenbergA . Deep neural networks in psychiatry [J]. Molecular Psychiatry, 2019. [本文引用: 1]
LiuX, ChenK, WuT, WeidmanD, LureF, LiJ . Use of multi-modality imaging and artificial intelligence for diagnosis and prognosis of early stages of alzheimer’s disease Transl Res.2018,194:56-67. [本文引用: 1]
LiuS, LiuS, CaiW, CheH, PujolS, KikinisR , et al. Multimodal neuroimaging feature learning for multiclass diagnosis of Alzheimer’s disease IEEE Trans Biomed Eng.2015,62:1132-40. [本文引用: 1]
MaJ, ZhangR . Automatic Calcium Scoring in Cardiac and Chest CT Using DenseRAUnet [J]. arXiv: Image and Video Processing, 2019. [本文引用: 1]
DouQ, ChenH, JinY, et al. 3D deeply supervised network for automatic liver segmentation from CT volumes [C]// International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2016: 149-157. [本文引用: 2]
GregorU, PriyamT, TalalA , et al. Deep Learning Localizes and Identifies Polyps in Real Time With 96% Accuracy in Screening Colonoscopy [J]. Gastroenterology, 2018: S0016508518346596. [本文引用: 2]
HavaeiM, DavyA, Warde-FarleyD , et al. Brain tumor segmentation with deep neural networks [J]. Medical image analysis, 2017,35:18-31. [本文引用: 1]
Menze BH, JakabA, BauerS , et al. The multimodal brain tumor image segmentation benchmark (BRATS) [J]. IEEE transactions on medical imaging, 2014,34(10):1993-2024. [本文引用: 1]
WangD, KhoslaA, GargeyaR , et al. Deep learning for identifying metastatic breast cancer [J]. arXiv preprint arXiv:1606.05718, 2016. [本文引用: 1]
GulshanV, PengL, CoramM , et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs [J]. Jama, 2016,316(22):2402-2410. [本文引用: 1]
Hansen MB, Abràmoff MichaelD, Folk JC , et al. Results of Automated Retinal Image Analysis for Detection of Diabetic Retinopathy from the Nakuru Study, Kenya [J]. PLOS ONE, 2015,10(10):e0139148. [本文引用: 1]
LiW, JiaF, HuQ . Automatic segmentation of liver tumor in CT images with deep convolutional neural networks [J]. Journal of Computer and Communications, 2015,3(11):146. [本文引用: 1]
ShenD, WuG, SukH , et al. Deep Learning in Medical Image Analysis [J]. Annual Review of Biomedical Engineering, 2017,19(1):221-248. [本文引用: 1]
SimonyanK, VedaldiA, ZissermanA, et al. Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps [C]. international conference on learning representations, 2013. [本文引用: 1]
LeeH, TajmirS, Lee JS , et al. Fully Automated Deep Learning System for Bone Age Assessment [J]. Journal of Digital Imaging, 2017,30(4):427-441. [本文引用: 1]
Erickson BJ, KorfiatisP, Kline TL , et al. Deep Learning in Radiology: Does One Size Fit All? [J]. Journal of The American College of Radiology, 2018,15(3):521-526. [本文引用: 1]
KaiserL, Gomez AN, ShazeerN , et al. One Model To Learn Them All [J]. arXiv: Learning, 2017. [本文引用: 1]