TY - JOUR
T1 - A Review of Deep Learning in Medical Imaging
T2 - Imaging Traits, Technology Trends, Case Studies with Progress Highlights, and Future Promises
AU - Zhou, S. Kevin
AU - Greenspan, Hayit
AU - Davatzikos, Christos
AU - Duncan, James S.
AU - Van Ginneken, Bram
AU - Madabhushi, Anant
AU - Prince, Jerry L.
AU - Rueckert, Daniel
AU - Summers, Ronald M.
N1 - Funding Information:
Manuscript received August 1, 2020; revised December 12, 2020; accepted January 13, 2021. Date of publication February 26, 2021; date of current version April 30, 2021. The work of Anant Madabhushi was supported in part by the National Institutes of Health under Award 1U24CA199374-01, Award R01CA202752-01A1, Award R01CA208236-01A1, Award R01CA216579-01A1, Award R01CA220581-01A1, Award 1U01CA239055-01, Award 1U54CA254566-01, Award 1U01CA248226-01, and Award 1R43EB028736-01 and in part by the VA Merit Review Award IBX004121A from the Biomedical Laboratory Research and Development Service of the United States Department of Veterans Affairs. The work of Hayit Greenspan was supported in part by the Israeli Science Foundation (ISF) and in part by the Ministry of Science & Technology. The work of Ronald M. Summers was supported by the National Institutes of Health Clinical Center. (S. Kevin Zhou and Hayit Greenspan contributed equally to this work.) (Corresponding author: S. Kevin Zhou.) S. Kevin Zhou is with the School of Biomedical Engineering, University of Science and Technology of China, Hefei 230052, China, and also with the Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China (e-mail: zhoushaohua@ict.ac.cn). Hayit Greenspan is with the Department of Biomedical Engineering, Faculty of Engineering, Tel-Aviv University, Tel-Aviv 69978, Israel. Christos Davatzikos is with the Radiology Department, University of Pennsylvania, Philadelphia, PA 19104 USA, and also with the Electrical and Systems Engineering Department, University of Pennsylvania, Philadelphia, PA 19104 USA. James S. Duncan is with the Department of Biomedical Engineering, Yale University, New Haven, CT 06520 USA, and also with the Department of Radiology and Biomedical Imaging, Yale University, New Haven, CT 06520 USA. Bram van Ginneken is with the Radboud University Medical Center, 6525 GA Nijmegen, The Netherlands. Anant Madabhushi is with the Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH 44106 USA, and also with the Louis Stokes Cleveland Veterans Administration Medical Center, Cleveland, OH 44106 USA. Jerry L. Prince is with the Electrical and Computer Engineering Department, Johns Hopkins University, Baltimore, MD 21218 USA. Daniel Rueckert is with the Klinikum rechts der Isar, Technical University of Munich (TU Munich), 81675 Munich, Germany, and also with the Department of Computing, Imperial College London, London SW7 2AZ, U.K. Ronald M. Summers is with the National Institutes of Health Clinical Center, Bethesda, MD 20892 USA.
Publisher Copyright:
© 1963-2012 IEEE.
PY - 2021/5
Y1 - 2021/5
N2 - Since its renaissance, deep learning (DL) has been widely used in various medical imaging tasks and has achieved remarkable success in many medical imaging applications, thereby propelling us into the so-called artificial intelligence (AI) era. It is known that the success of AI is mostly attributed to the availability of big data with annotations for a single task and the advances in high-performance computing. However, medical imaging presents unique challenges that confront DL approaches. In this survey article, we first present traits of medical imaging, highlight both clinical needs and technical challenges in medical imaging, and describe how emerging trends in DL are addressing these issues. We cover the topics of network architecture, sparse and noisy labels, federating learning, interpretability, uncertainty quantification, and so on. Then, we present several case studies that are commonly found in clinical practice, including digital pathology and chest, brain, cardiovascular, and abdominal imaging. Rather than presenting an exhaustive literature survey, we instead describe some prominent research highlights related to these case study applications. We conclude with a discussion and presentation of promising future directions.
AB - Since its renaissance, deep learning (DL) has been widely used in various medical imaging tasks and has achieved remarkable success in many medical imaging applications, thereby propelling us into the so-called artificial intelligence (AI) era. It is known that the success of AI is mostly attributed to the availability of big data with annotations for a single task and the advances in high-performance computing. However, medical imaging presents unique challenges that confront DL approaches. In this survey article, we first present traits of medical imaging, highlight both clinical needs and technical challenges in medical imaging, and describe how emerging trends in DL are addressing these issues. We cover the topics of network architecture, sparse and noisy labels, federating learning, interpretability, uncertainty quantification, and so on. Then, we present several case studies that are commonly found in clinical practice, including digital pathology and chest, brain, cardiovascular, and abdominal imaging. Rather than presenting an exhaustive literature survey, we instead describe some prominent research highlights related to these case study applications. We conclude with a discussion and presentation of promising future directions.
KW - Deep learning (DL)
KW - medical imaging
KW - survey
UR - http://www.scopus.com/inward/record.url?scp=85101829337&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85101829337&partnerID=8YFLogxK
U2 - 10.1109/JPROC.2021.3054390
DO - 10.1109/JPROC.2021.3054390
M3 - Article
AN - SCOPUS:85101829337
SN - 0018-9219
VL - 109
SP - 820
EP - 838
JO - Proceedings of the IEEE
JF - Proceedings of the IEEE
IS - 5
M1 - 9363915
ER -