浏览全部资源
扫码关注微信
Received：09 July 2020，
Revised：2020-10-18，
Accepted：25 October 2020，
Published：16 March 2021
移动端阅览
医学大数据主要包括电子健康档案数据（electronic health record，EHR）、医学影像数据和基因信息数据等，其中医学影像数据占现阶段医学数据的绝大部分。如何将医学大数据应用于临床实践?这是计算机科学研究人员非常关注的问题，医学人工智能提供了一个很好的答案。通过结合医学图像大数据分析方向截至2020年的最新研究进展，以及医学图像大数据分析领域最近的工作，梳理了当前在医学图像领域以核磁共振影像、超声影像、病理和电信号为代表的4个子领域以及部分其他方向使用深度学习进行图像分析的方法理论和主要流程，对不同算法进行结果评价。本文分析了现有算法的优缺点以及医学影像领域的重难点，介绍了智能成像和深度学习在大数据分析以及疾病早期诊断领域的应用，同时展望了本领域未来的发展热点。深度学习在医学影像领域发展迅速，发展前景广阔，对疾病的早期诊断有重要作用，能有效提高医生工作效率并减轻负担，具有重要的理论研究和实际应用价值。
Medical big data mainly include electronic health record data
such as medical imaging data and genetic information data
among which medical imaging data takes up the most of medical data currently. One of the problems that researchers in computer science are greatly concerned about is how to apply medical big data in clinical practice.Artificial intelligence (AI) provides a good way to address this problem. AI algorithms
particularly deep learning
have demonstrated remarkable progress in image-recognition tasks. Historically
in radiology practice
trained physicians visually assess medical images for the detection
characterization
and monitoring of diseases. AI methods excel at automatically recognizing complex patterns in imaging data and providing quantitative
rather than qualitative
assessments of radiographic characteristics. Methods ranging from convolutional neural networks to variational autoencoders have found myriad applications in the medical image analysis field
propelling it forward at a rapid pace. In this review
by combining recent work and the latest research progress of big data analysis of medical images until 2020
we have summarized the theory
main process
and evaluation results of multiple deep learning algorithms in some fields of medical image analysis
including magnetic resonance imaging (MRI)
pathology imaging
ultrasound imaging
electrical signals
digital radiography
molybdenum target
and diabetic eye imaging
using deep learning. MRI is one of the main research areas of medical image analysis. The existing research literature includes Alzheimer's disease MRI
Parkinson's disease MRI
brain tumor MRI
prostate cancer MRI
and cardiac MRI. MRI is also divided into two-dimensional and three-dimensional image analysis
especially for three-dimensional data
where insufficient data volume leads to problems such as overfitting
large calculations
and slow training.Medical ultrasound (also known as diagnostic sonography or ultrasonography) is a diagnostic imaging technique or therapeutic application of ultrasound. It is used to create an image of internal body structures such as tendons
muscles
joints
blood vessels
and internal organs. It aims to find the source of a disease or to exclude pathology. The practice of examining pregnant women using ultrasound is called obstetric ultrasonography and was an early development and application of clinical ultrasonography.Ultrasonography uses sound waves with higher frequencies than those audible to humans (>20 000 Hz). Ultrasonic images
also known as sonograms
are made by sending ultrasound pulses into the tissue using a probe. The ultrasound pulses echo off tissues with different reflection properties and are recorded and displayed as an image.Many different types of images can be formed. The most common is a B-mode image (brightness)
which displays the acoustic impedance of a two-dimensional cross-section of a tissue. Other types can display blood flow
tissue motion over time
the location of blood
the presence of specific molecules
the stiffness of a tissue
or the anatomy of a three-dimensional region. Pathology is the gold standard for diagnosing some diseases
especially digital image of pathology.We specifically discuss AI combined with digital pathology images for diagnosis.Electroencephalography (EEG) is an electrophysiological monitoring method to record the electrical activity of the brain. It is typically noninvasive
with the electrodes placed along the scalp.However
invasive electrodes are sometimes used
for example in electrocorticography
sometimes called intracranial EEG. EEG is most often used to diagnose epilepsy
which causes abnormalities in EEG readings. It is also used to diagnose sleep disorders
depth of anesthesia
coma
encephalopathies
and brain death. EEG used to be a first-line method of diagnosis for tumors
stroke
and other focal brain disorders
but its use has decreased with the advent of high-resolution anatomical imaging techniques such as MRI and computed tomography (CT). Despite limited spatial resolution
EEG continues to be a valuable tool for research and diagnosis. It is one of the few mobile techniques available and offers millisecond-range temporal resolution
which is not possible with CT
positron emission tomography (PET)
or MRI.Electrocardiography(ECG or EKG) is the process of producing an electrocardiogram. It is a graph of voltage versus time of the electrical activity of the heart using electrodes placed on the skin. These electrodes detect small electrical changes that are a consequence of cardiac muscle depolarization followed by repolarization during each cardiac cycle (heartbeat). Changes in the normal ECG pattern occur in numerous cardiac abnormalities
including cardiac rhythm disturbances (e.g.
atrial fibrillation and ventricular tachycardia)
inadequate coronary artery blood flow (e.g.
myocardial ischemia and myocardial infarction)
and electrolyte disturbances (e.g.
hypokalemia and hyperkalemia).We analyzed the advantages and disadvantages of existing algorithms and the important and difficult points in the field of medical imaging
and introduced the application of intelligent imaging and deep learning in the field of big data analysis and early disease diagnosis. The current algorithms in the field of medical imaging have made considerable progress
but there is still a lot of room for development. We also focus on the optimization and improvement of different algorithms in different sub-fields under a variety of segmentation and classification indicators (e.g.
Dice
IoU
accuracy and recall rate)
and we look forward to the future development hotspots in this field. Deep learning has developed rapidly in the field of medical imaging and has broad prospects for development. It plays an important role in the early diagnosis of diseases. It can effectively improve the work efficiency of doctors and reduce their burden. Moreover
it has important theoretical research and practical application value.
Cai L, Gao J Y and Zhao D. 2020. A review of the application of deep learning in medical image classification and segmentation. Annals of Translational Medicine, 8(11): #713[DOI: 10.21037/atm.2020.02.44]
Chen L, Cui Y T, Song H, Huang B X, Yang J, Zhao D and Xia B. 2020. Femoral head segmentation based on improved fully convolutional neural network for ultrasound images. Signal, Image and Video Processing, 14(5): 1043-1051[DOI: 10.1007/s11760-020-01637-z]
Chen S W, Liu Y J, Liu D, Su C, Zhao D, Qian L X and Zhang P H. 2019. Alex model and adaptive contrast enhancement based ultrasound imaging classification. Computer Science, 46(6A): 146-152
陈思文, 刘玉江, 刘冬, 苏晨, 赵地, 钱林学, 张佩珩. 2019. 基于AlexNet模型和自适应对比度增强的乳腺结节超声图像的分类. 计算机科学, 46(6A): 146-152 http://www.jsjkx.com/CN/abstract/abstract18284.shtml
Gao Q, Gao J Y and Zhao D. 2020. GNNI U-Net: precise segmentation neural network of left ventricular contours for MRI images based on group normalization and nearest interpolation. Computer Science, 47(8): 213-220
高强, 高敬阳, 赵地. 2020. GNNI U-Net: 基于组归一化与最近邻插值的MRI左心室轮廓精准分割网络. 计算机科学, 47(8): 213-220[DOI: 10.11896/jsjkx.190600026]
Guo X J, Zeng C, Sun S Y, Zhao D and Hua Z Q. 2020. QRS complex detection algorithm based on 1-D CNN. Computer Engineering and Design, 41(9): 2469-2475
郭新杰, 曾成, 孙尚云, 赵地, 花中秋. 2020. 基于1-D CNN的QRS波群检测算法. 计算机工程与设计, 41(9): 2469-2475
Guo Z H, Zhou J Q and Zhao D. 2020. Thyroid nodule ultrasonic imaging segmentation based on a deep learning model and data augmentation//Proceedings of the 4th IEEE Information Technology, Networking, Electronic and Automation Control Conference, Chongqing, China: IEEE: 549-554
Li L W, Xiong M, Qian L X, Zhao D, Chen J J and Chi X B. 2018. Bultrasound imaging diagnosis of fatty liver based on adaptive contrast enhancement and deep CNN. Computer Application Research, 35: 402-404, 293
李丽玮, 熊鸣, 钱林学, 赵地, 陈金军, 迟学斌. 2018. 基于自适应对比度增强和深度CNN的脂肪肝B超影像诊断. 计算机应用研究, 35: 402-404, 293
Lyu H M, Zhao D and Chi X B. 2017. Deep learning for early diagnosis of Alzheimer's disease based on intensive AlexNet. Computer Science, 44(6A): 50-53
吕鸿蒙, 赵地, 迟学斌. 2017. 基于增强AlexNet的深度学习的阿尔茨海默病的早期诊断. 计算机科学, 44(6A): 50-53)[ DOI: CNKI:SUN:JSJA.0.2017-S1-011 http://dx.doi.org/CNKI:SUN:JSJA.0.2017-S1-011 ]
Su Q H, Fu J C, Gu H, Zhang S S, Li Y F, Jiang F Z, Bai H L and Zhao D. 2019a. Parallel algorithm design for assisted diagnosis of prostate cancer. Computer Science, 46(11A): 524-527
苏庆华, 付景超, 谷焓, 张姗姗, 李奕飞, 江方舟, 白翰林, 赵地. 2019a. 前列腺癌辅助诊断GPU并行算法设计. 计算机科学, 46(11A): 524-527
Su Q H, Zhang S S, Cai L, Gu H, Li Y F, Yu G H, Jiang F Z, Bai H L and Zhao D. 2019b. Prostate diagnostic model based on 3D classification network. China Digital Medicine, 14(3): 18-21
苏庆华, 张珊珊, 蔡磊, 谷焓, 李奕飞, 俞戈浩, 江方舟, 白翰林, 赵地. 2019b. 基于三维分类网络的前列腺辅助诊断. 中国数字医学, 14(3): 18-21[DOI: 10.3969/j.issn.1673-7571.2019.03.005]
Sun Z Y, Shi X L, Zhu Y W, Zhang S S, Guo J F and Zhao D. 2019. Mammary molybdenum target screening based on deep convolutional neural network. China Digital Medicine, 14(3): 62-65
孙泽宇, 史晓林, 朱延武, 张珊珊, 郭建飞, 赵地. 2019. 基于深度卷积神经网络的乳腺钼靶筛查. 中国数字医学, 14(3): 62-65[DOI: 10.3969/j.issn.1673-7571.2019.03.002]
Wang H Z, Zhao D, Yang L Q, Xia T, Zhou X Y and Miao Z Y. 2018. An approach for training data enrichment and batch labeling in AI+MRI aided diagnosis. Chinese Journal of Magnetic Resonance, 35(4): 447-456
汪红志, 赵地, 杨丽琴, 夏天, 周皛月, 苗志英. 2018. 基于AI+MRI的影像诊断的样本增广与批量标注方法. 波谱学杂志, 35(4): 447-456[DOI: 10.11938/cjmr20182658]
Wang J X, Gao D, Zhao D, Zhou L X, Geng Z J, Zong H Q, Cao H Z and Xie Z W. 2019. Study on the aging of resting functional connections in vermis of nornal human brain. Hebei Medical Jounary, 41(20): 3052-3056
王佳鑫, 高铎, 赵地, 周立霞, 耿左军, 宗会迁, 曹恒志, 谢志伟. 2019. 正常人小脑蚓部静息态功能连接的脑老化研究. 河北医药, 41(20): 3052-3056[DOI: 10.3969/j.issn.1002-7386.2019.20.002]
Wei X N, Xing J Q, Wang Z Y, Wang Y S, Shi J, Zhao D and Wang H Z. 2020. Magnetic resonance image segmentation of articular synovium based on improved U-Net. Journal of Computer Applications, 40(11): 3340-3345
魏小娜, 邢嘉祺, 王振宇, 王颖珊, 石洁, 赵地, 汪红志. 2020. 基于改进U-Net网络的关节滑膜磁共振图像的分割. 计算机应用, 40(11): 3340-3345)[ DOI: 1001-9081(2020)11-3340-06 http://dx.doi.org/1001-9081(2020)11-3340-06 ]
Yan S L, Lin Y X, Li H X, Zhao D and Chi X B. 2019. Diabetic retinopathy detection based on multiple transfer learning. China Digital Medicine, 14(3): 26-30
颜嵩林, 林溢星, 李鹤喜, 赵地, 迟学斌. 2019. 基于多重迁移学习的糖尿病视网膜病变检测, 中国数字医学, 14(3): 26-30[DOI: 10.3969/j.issn.1673-7571.2019.03.007]
Yang T, Zhang S S, Jiang F Z, Li Y F, Yu G H and Zhao D. 2019. Brachial plexus ultrasound image optimization based on deep learning and adaptive contrast enhancement. Computer Science, 46(11A): 236-240
杨桐, 张珊珊, 江方舟, 李奕飞, 俞戈昊, 赵地. 2019. 基于深度学习与自适应对比度增强的臂丛神经超声图像优化, 计算机科学, 46(11A): 236-240
Yang Y, Yan L F, Zhang X, Han Y, Nan H Y, Hu Y C, Hu B, Yan S L, Zhang J, Cheng D L, Ge X W, Cui G B, Zhao D and Wang W. 2018. Glioma grading on conventional MR images: a deep learning study with transfer learning. Front Neurosci, 12: #804[DOI: 10.3389/fnins.2018.00804]
Yao T C, Xiao L, Zhao D and Sun Y Z. 2018. GPU computing based fast discrete wavelet transform for l1-regularized SPIRiT reconstruction. The Imaging Science Journal, 66(7): 393-408[DOI: 10.1080/13682199.2018.1496220]
Yao T C, Zhao D and Chi X B. 2017. Study on early warning of Alzheimer's disease based on multiple GPU and convolutional neural network. CCF HPC China
姚铁锤, 赵地, 迟学斌. 2017. 基于多GPU计算与卷积神经网络的阿尔茨海默症早期预警研究. HPC China 2017
Zhang J H, Han X, Zhao H W, Zhao D, Wang N, Zhao T, He G N, Zhu X R, Zhang Y, Han J Y and Huang D L. 2018. Personalized prediction model for seizure-free epilepsy with levetiracetam therapy: a retrospective data analysis using support vector machine. British Journal of Clinical Pharmacology, 84(11): 2615-2624[DOI: 10.1111/bcp.13720]
Zhang Q L, Chi X B and Zhao D. 2018. Early diagnosis of parkinson's disease based on deep learning. Computer Systems and Applications, 27(9): 1-9
张巧丽, 迟学斌, 赵地. 2018. 基于深度学习的帕金森病症早期诊断. 计算机系统应用, 27(9): 1-9[DOI: 10.15888/j.cnki.csa.006538]
Zhang Q L, Zhao D and Chi X B. 2017. Review for deep learning based on medical lmaging diagnosis. Computer Science, 44(11A): 1-7[DOI: 10.11896/j.issn.1002-137X.2017.11A.001]
Zhang Z Z, Gao J Y, Lyu G and Zhao D. 2018. Pathological image classification of gastric cancer based on depth learning. Computer Science, 45(11A): 263-268
张泽中, 高敬阳, 吕纲, 赵地. 2018. 基于深度学习的胃癌病理图像分类方法. 计算机科学, 45(11A): 263-268[DOI: 10.11896/j.issn.1002-137X.2017.11A.001]
Zhang Z Z, Gao J Y and Zhao D. 2019. MIFNet: pathological image segmentation method for stomach cancer based on multi-scale input and feature fusion. Journal of Computer Applications, 39(S2): 107-113
张泽中, 高敬阳, 赵地. 2019. MIFNet: 基于多尺度输入与特征融合的胃癌病理图像分割方法. 计算机应用, 39(S2): 107-113)[ DOI: 1001-9081(2019)S2-0107-07 http://dx.doi.org/1001-9081(2019)S2-0107-07 ]
相关文章
相关作者
相关机构
京公网安备11010802024621
