Strategies for Implementing Machine Learning Algorithms in the Clinical Practice of Radiology

被引:13
作者
Chae, Allison [5 ]
Yao, Michael S. [1 ,5 ]
Sagreiya, Hersh [2 ,5 ]
Goldberg, Ari D. [6 ]
Chatterjee, Neil [2 ]
MacLean, Matthew T. [2 ]
Duda, Jeffrey [2 ]
Elahi, Ameena [7 ]
Borthakur, Arijitt [2 ,5 ,8 ]
Ritchie, Marylyn D. [3 ]
Rader, Daniel [4 ]
Kahn, Charles E. [2 ,5 ]
Witschey, Walter R. [2 ,5 ]
Gee, James C. [2 ,5 ]
机构
[1] Univ Penn, Dept Bioengn, 3400 Civ Ctr Blvd, Philadelphia, PA 19104 USA
[2] Univ Penn, Dept Radiol, 3400 Civ Ctr Blvd, Philadelphia, PA 19104 USA
[3] Univ Penn, Dept Genet, 3400 Civ Ctr Blvd, Philadelphia, PA 19104 USA
[4] Univ Penn, Dept Med, 3400 Civ Ctr Blvd, Philadelphia, PA 19104 USA
[5] Univ Penn, Perelman Sch Med, 3400 Civ Ctr Blvd, Philadelphia, PA 19104 USA
[6] Loyola Univ, Med Ctr, Dept Radiol, Maywood, IL USA
[7] Univ Penn, Dept Informat Serv, Philadelphia, PA USA
[8] Univ Penn, Leonard Davis Inst Hlth Econ, Philadelphia, PA USA
基金
美国国家卫生研究院;
关键词
ARTIFICIAL-INTELLIGENCE; BIOBANK; QUANTIFICATION;
D O I
10.1148/radiol.223170
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
Despite recent advancements in machine learning (ML) applications in health care, there have been few benefits and improvements to clinical medicine in the hospital setting. To facilitate clinical adaptation of methods in ML, this review proposes a standardized framework for the step-by-step implementation of artificial intelligence into the clinical practice of radiology that focuses on three key components: problem identification, stakeholder alignment, and pipeline integration. A review of the recent literature and empirical evidence in radiologic imaging applications justifies this approach and offers a discussion on structuring implementation efforts to help other hospital practices leverage ML to improve patient care. © 2024 Radiological Society of North America Inc.. All rights reserved.
引用
收藏
页数:12
相关论文
共 109 条
[92]   Open-Source MR Imaging and Reconstruction Workflow [J].
Veldmann, Marten ;
Ehses, Philipp ;
Chow, Kelvin ;
Nielsen, Jon-Fredrik ;
Zaitsev, Maxim ;
Stoecker, Tony .
MAGNETIC RESONANCE IN MEDICINE, 2022, 88 (06) :2395-2407
[93]   A synthesis of implementation science frameworks and application to global health gaps [J].
Villalobos Dintrans, Pablo ;
Bossert, Thomas J. ;
Sherry, Jim ;
Kruk, Margaret E. .
GLOBAL HEALTH RESEARCH AND POLICY, 2019, 4 (01)
[94]   Machine Learning Model Cards Transparency Review Using model card toolkit [J].
Wadhwani, Abhishek ;
Jain, Priyank .
2020 IEEE PUNE SECTION INTERNATIONAL CONFERENCE (PUNECON), 2020, :133-137
[95]  
Wei JS, 2022, Arxiv, DOI arXiv:2201.11903
[96]  
Wiley L, 2022, medRxiv
[97]   Toward Foundational Deep Learning Models for Medical Imaging in the New Era of Transformer Networks [J].
Willemink, Martin J. ;
Roth, Holger R. ;
Sandfort, Veit .
RADIOLOGY-ARTIFICIAL INTELLIGENCE, 2022, 4 (06)
[98]   Preparing Medical Imaging Data for Machine Learning [J].
Willemink, Martin J. ;
Koszek, Wojciech A. ;
Hardell, Cailin ;
Wu, Jie ;
Fleischmann, Dominik ;
Harvey, Hugh ;
Folio, Les R. ;
Summers, Ronald M. ;
Rubin, Daniel L. ;
Lungren, Matthew P. .
RADIOLOGY, 2020, 295 (01) :4-15
[99]   Improving breast cancer diagnostics with deep learning for MRI [J].
Witowski, Jan ;
Heacock, Laura ;
Reig, Beatriu ;
Kang, Stella K. ;
Lewin, Alana ;
Pysarenko, Kristine ;
Patel, Shalin ;
Samreen, Naziya ;
Rudnicki, Wojciech ;
Luczynska, Elzbieta ;
Popiela, Tadeusz ;
Moy, Linda ;
Geras, Krzysztof J. .
SCIENCE TRANSLATIONAL MEDICINE, 2022, 14 (664)
[100]   Self-supervised learning of physics-guided reconstruction neural networks without fully sampled reference data [J].
Yaman, Burhaneddin ;
Hosseini, Seyed Amir Hossein ;
Moeller, Steen ;
Ellermann, Jutta ;
Ugurbil, Kamil ;
Akcakaya, Mehmet .
MAGNETIC RESONANCE IN MEDICINE, 2020, 84 (06) :3172-3191