Advertisement

The Role of an Artificial Intelligence Ecosystem in Radiology

  • Bibb Allen
  • Robert Gish
  • Keith Dreyer
Chapter

Abstract

Moving artificial intelligence tools for diagnostic imaging into routine clinical practice will require cooperation and collaboration between developers, physicians, regulators, and health system administrators. Radiologists can play an important role in promoting this AI ecosystem by delineating AI use cases for diagnostic imaging and developing standardized data elements and workflow integration interfaces. Structured AI use cases that define specific parameters for datasets for algorithm training and testing can promote multiple sites to develop training, and validation datasets, which can help ensure algorithms respect technical, geographic, and demographic diversity in patient populations and image acquisition, are free of unintended bias and are generalizable to widespread clinical practice. Medial specialty societies can play a role in protecting patients from unintended consequences of AI through use case development and developing programs for independent algorithm validation and monitoring the effectiveness and safety of AI tools in clinical practice through AI registries. The development and implementation of AI algorithms for medical imaging will benefit from the establishment of an AI ecosystem that includes physicians, researchers, software developers along with governmental regulatory agencies, the HIT industry, and hospital administrators all working to bring AI tools safely and efficiently into clinical practice.

Keywords

Artificial intelligence ecosystem Artificial intelligence use case Artificial intelligence government regulation Artificial intelligence data registries Artificial intelligence common data elements 

References

  1. 1.
    Allen B, Dreyer K. The artificial intelligence ecosystem for the radiological sciences: ideas to clinical practice. J Am Coll Radiol. 2018;  https://doi.org/10.1016/j.jacr.2018.02.032.CrossRefGoogle Scholar
  2. 2.
    JASON 2017. Artificial intelligence for health and heath care. JSR-17-Task-002.Google Scholar
  3. 3.
    Definition of Ecosystem. [Internet]. Merrian-webster.com. 2018 [cited 10 June 2018]. Available from: https://www.merriam-webster.com/dictionary/ecosystem
  4. 4.
    Moore JF. Predators and prey: a new ecology of competition. Harv Bus Rev. 1993 May 1;71(3):75–86.PubMedGoogle Scholar
  5. 5.
    Moore JF. The death of competition: leadership and strategy in the age of business ecosystems. New York: HarperBusiness; 1996 May.Google Scholar
  6. 6.
    Messerschmitt DG, Szyperski C. Software ecosystem: understanding an indispensable technology and industry, vol. 1. London: MIT Press Books; 2005.CrossRefGoogle Scholar
  7. 7.
    Seddon JJ, Currie WL. Cloud computing and trans-border health data: unpacking US and EU healthcare regulation and compliance. Health Policy Technol. 2013 Dec 1;2(4):229–41.CrossRefGoogle Scholar
  8. 8.
    Barnett, JC, Berchick, ER. Current population reports, P60–260, Health Insurance Coverage in the United States: 2016, U.S. Washington, DC: Government Printing Office; 2017.Google Scholar
  9. 9.
    Berwick DM, Nolan TW, Whittington J. The triple aim: care, health, and cost. Health Aff. 2008 May;27(3):759–69.CrossRefGoogle Scholar
  10. 10.
    Bodenheimer T, Sinsky C. From triple to quadruple aim: care of the patient requires care of the provider. Ann Fam Med. 2014 Nov 1;12(6):573–6.CrossRefGoogle Scholar
  11. 11.
    Sikka R, Morath JM, Leape L. The Quadruple Aim: care, health, cost and meaning in work. BMJ Qual Saf.  https://doi.org/10.1136/bmjqs-2015-004160.CrossRefGoogle Scholar
  12. 12.
    Thrall JH, Li X, Li Q, Cruz C, Do S, Dreyer K, Brink J. Artificial intelligence and machine learning in radiology: opportunities, challenges, pitfalls, and criteria for success. J Am Coll Radiol. 2018 Mar 1;15(3):504–8.CrossRefGoogle Scholar
  13. 13.
    Lakhani P, Prater AB, Hutson RK, Andriole KP, Dreyer KJ, Morey J, Prevedello LM, Clark TJ, Geis JR, Itri JN, Hawkins CM. Machine learning in radiology: applications beyond image interpretation. J Am Coll Radiol. 2017 Nov 17;15(2):350–9.CrossRefGoogle Scholar
  14. 14.
    Erdal BS, Prevedello LM, Qian S, Demirer M, Little K, Ryu J, O’Donnell T, White RD. Radiology and Enterprise Medical Imaging Extensions (REMIX). J Digit Imaging. 2018 Feb 1;31(1):91–106. CrossRefGoogle Scholar
  15. 15.
    Huffman J. Healthcare Information and Management Systems Society. 2018 March 6.Google Scholar
  16. 16.
    Turing AM. Computing machinery and intelligence. Mind. 1950 Oct;59(236):433.CrossRefGoogle Scholar
  17. 17.
    Minsky M. Steps toward artificial intelligence. Proc IRE. 1961 Jan;49(1):8–30.CrossRefGoogle Scholar
  18. 18.
    McCarthy J. From here to human-level AI. In Proc. of principles of knowledge representation and reasoning (KR 1996).Google Scholar
  19. 19.
    Taubes G. The rise and fall of thinking machines. Inc. 1995;17(13):61–5.Google Scholar
  20. 20.
    Yang Z, Zhu Y, Pu Y. Parallel image processing based on CUDA. In Computer Science and Software Engineering, 2008 International Conference on 2008 Dec 12 (vol. 3, pp. 198–201). IEEE.Google Scholar
  21. 21.
    Ciregan D, Meier U, Schmidhuber J. Multi-column deep neural networks for image classification. In Computer vision and pattern recognition (CVPR), 2012 IEEE conference on 2012 Jun 16 (pp. 3642–3649). IEEE.Google Scholar
  22. 22.
    Mobile Fact Sheet. Pew Research Center: Internet, Science & Tech. 2018 [cited 10 June 2018]. Available from http://www.pewinternet.org/fact-sheet/mobile/
  23. 23.
    Chockley K, Emanuel E. The end of radiology? Three threats to the future practice of radiology. J Am Coll Radiol. 2016 Dec 1;13(12):1415–20.CrossRefGoogle Scholar
  24. 24.
    Remnick D. Obama reckons with a Trump presidency. The New Yorker. 2016 Nov;28:28.Google Scholar
  25. 25.
    Hinton G. Geoff Hinton on Radiology. Machine Learning and Market for Intelligence Conference, Creative Disruption Lab Toronto, Canada. 2016. Viewable at: https://www.youtube.com/watch?v=2HMPRXstSvQ
  26. 26.
    Oncology Expert Advisor [Internet]. MD Anderson Cancer Center. 2018 [cited 10 June 2018]. Available from: https://www.mdanderson.org/publications/annual-report/annual-report-2013/the-oncology-expert-advisor.html
  27. 27.
    Herper M. MD Anderson benches IBM Watson in setback for artificial intelligence in medicine. Forbes. Zugriff im Juli. 2017 Feb.Google Scholar
  28. 28.
    Ravì D, Wong C, Deligianni F, Berthelot M, Andreu-Perez J, Lo B, Yang GZ. Deep learning for health informatics. IEEE J Biomed Health Inform. 2017 Jan;21(1):4–21.CrossRefGoogle Scholar
  29. 29.
    Deo RC. Machine learning in medicine. Circulation. 2015 Nov 17;132(20):1920–30.CrossRefGoogle Scholar
  30. 30.
    Valente IR, Cortez PC, Neto EC, Soares JM, de Albuquerque VH, Tavares JM. Automatic 3D pulmonary nodule detection in CT images: a survey. Comput Methods Programs Biomed. 2016 Feb 1;124:91–107.CrossRefGoogle Scholar
  31. 31.
    Spampinato C, Palazzo S, Giordano D, Aldinucci M, Leonardi R. Deep learning for automated skeletal bone age assessment in X-ray images. Med Image Anal. 2017 Feb 1;36:41–51.CrossRefGoogle Scholar
  32. 32.
    Buolamwini J, Gebru T. Gender shades: Intersectional accuracy disparities in commercial gender classification. In Conference on Fairness, Accountability and Transparency 2018 Jan 21 (pp. 77–91).Google Scholar
  33. 33.
    Health Insurance Portability and Accountability Act of 1996 (HIPAA.)Pub. L. 104–191, 110 Stat. 1936 (1996)Google Scholar
  34. 34.
    The HIPAA Privacy Rule. 45 CFR 160, 162, and 164. 28 Dec 2000.Google Scholar
  35. 35.
    The Security Rule. 45 CFR Part 160 and Subparts A and C of Part 164. 20 Feb 2003.Google Scholar
  36. 36.
  37. 37.
    AI has no place in the NHS If patient privacy isn’t assured. Wired. http://www.wired.co.uk/article/ai-healthcare-gp-deepmind-privacy-problems
  38. 38.
    US Food and Drug Administration. What we do. https://www.fda.gov/AboutFDA/WhatWeDo/
  39. 39.
    US Food and Drug Administration. Medical Devices.Google Scholar
  40. 40.
    The 21st Century Cures Act. Pub. L. 114–255.Google Scholar
  41. 41.
  42. 42.
    US Food and Drug Administration. Software as a medical device. Do. https://www.fda.gov/MedicalDevices/DigitalHealth/SoftwareasaMedical Device/default.htm
  43. 43.
    US Food and Drug Administration. International Medical Device Regulators Forum. https://www.fda.gov/MedicalDevices/International Programs/IMDRF/default.htm
  44. 44.
  45. 45.
    US Food and Drug Administration. Medical Device Development Tools Program. https://www.fda.gov/MedicalDevices/ScienceandResearch/MedicalDevi ceDevelopmentToolsMDDT
  46. 46.
    US Food and Drug Administration. National Evaluation System for Health Technology. https://www.fda.gov/aboutfda/centersoffices/office ofmedicalproductsandtobacco/cdrh/cdrhreports/ucm301912.htm
  47. 47.
    US Food and Drug Administration. National evaluation system for health technology demonstration projects. https://nestcc.org/demonstration-projects/
  48. 48.
    Lund-RADS Assist: Advanced radiology guidance, reporting and monitoring. https://www.acr.org/Media-Center/ACR-News-Releases/2018/FDA-NEST-Program-Names-ACR-DSI-Use-Case-as-Demo-Project
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
    US FDA de novo approval clinical decision support software for stroke. https://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm596575.htm
  54. 54.
    US FDA de novo approval artificial intelligence based device to detect diabetes related eye problems. https://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm604357.htm
  55. 55.
    US FDA de novo approval of artificial intelligence algorithm for aiding providers in detecting wrist fractures. https://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm608833.htm
  56. 56.
  57. 57.
  58. 58.
    USFDA approval QuantX as Class II device. https://www.accessdata.fda.gov/cdrh_docs/pdf17/DEN170022.pdf
  59. 59.
    Boland GW, Duszak R, McGinty G, Allen B. Delivery of appropriateness, quality, safety, efficiency and patient satisfaction. J Am Coll Radiol. 2014 Jan 1;11(1):7–11.CrossRefGoogle Scholar
  60. 60.
  61. 61.
  62. 62.
    LOINC. Available at: http://loinc.org/about/
  63. 63.
    Alkasab TK, Bizzo BC, Berland LL, Nair S, Pandharipande PV, Harvey HB. Creation o an open framework for point-of-care computer-assisted reporting and decision support tools for radiologists. J Am Coll Radiol. 2017 Sep 1;14(9):1184–9.CrossRefGoogle Scholar
  64. 64.
    A Brief History of DICOM. In: Digital Imaging and Communications in Medicine (DICOM). Berlin, Heidelberg: Springer; 2008.Google Scholar
  65. 65.
    HL7 protocols. http://www.hl7.org
  66. 66.
    Fast Healthcare Interoperability Resources Specification. http://www.hl7.org/implement/standards/product_brief.cfm?product_id=449
  67. 67.
    Rubin DL, Kahn CE Jr. Common data elements in radiology. Radiology. 2016 Nov 10;283(3):837–44.CrossRefGoogle Scholar
  68. 68.
    Winget MD, Baron JA, Spitz MR, Brenner DE, Warzel D, Kincaid H, Thornquist M, Feng Z. Development of common data elements: the experience of and recommendations from the early detection research network. Int J Med Inform. 2003 Apr 1;70(1):41–8.CrossRefGoogle Scholar
  69. 69.
    Morin RL, Coombs LP, Chatfield MB. ACR dose index registry. J Am Coll Radiol. 2011 Apr 1;8(4):288–91.CrossRefGoogle Scholar
  70. 70.
    ACR National Radiology Data Registry. https://nrdr.acr.org/Portal/Nrdr/Main/page.aspx
  71. 71.
    Langlotz CP. RadLex: a new method for indexing online educational materials. Radiographics. 2006;26(6)CrossRefGoogle Scholar
  72. 72.
    Structured Reporting. http://www.radreport.org
  73. 73.
  74. 74.
    Boland GW, Thrall JH, Gazelle GS, Samir A, Rosenthal DI, Dreyer KJ, Alkasab TK. Decision support for radiologist report recommendations. J Am Coll Radiol. 2011 Dec 1;8(12):819–23.CrossRefGoogle Scholar
  75. 75.
  76. 76.
    Miller T, Howe P, Sonenberg L. Explainable AI: Beware of inmates running the asylum. InIJCAI-17 Workshop on Explainable AI (XAI). 2017 (p. 36).Google Scholar
  77. 77.
  78. 78.
    Luo W, Phung D, Tran T, Gupta S, Rana S, Karmakar C, Shilton A, Yearwood J, Dimitrova N, Ho TB, Venkatesh S. Guidelines for developing and reporting machine learning predictive models in biomedical research: a multidisciplinary view. J Med Internet Res. 2016 Dec;18(12)CrossRefGoogle Scholar
  79. 79.
    Stodden V. Reproducible research for scientific computing: Tools and strategies for changing the culture. Comput Sci Eng. 2012 Jul;14(4):13–7.CrossRefGoogle Scholar
  80. 80.
  81. 81.
    Iglovikov V, Rakhlin A, Kalinin A, Shvets A. Pediatric Bone Age Assessment Using Deep Convolutional Neural Networks. arXiv preprint arXiv:1712.05053. 2017 Dec 13.Google Scholar
  82. 82.
  83. 83.
    Rajpurkar P, Irvin J, Zhu K, Yang B, Mehta H, Duan T, Ding D, Bagul A, Langlotz C, Shpanskaya K, Lungren MP. CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning. arXiv preprint arXiv:1711.05225. 2017 Nov 14.Google Scholar
  84. 84.
    Gulshan V, Peng L, Coram M, Stumpe MC, Wu D, Narayanaswamy A, Venugopalan S, Widner K, Madams T, Cuadros J, Kim R. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA. 2016 Dec 13;316(22):2402–10.CrossRefGoogle Scholar
  85. 85.
    Esteva A, Kuprel B, Novoa RA, Ko J, Swetter SM, Blau HM, Thrun S. Dermatologist-level classification of skin cancer with deep neural networks. Nature. 2017 Feb;542(7639):115.CrossRefGoogle Scholar
  86. 86.
  87. 87.
  88. 88.
  89. 89.
    RSNA Machine Learning Showcase. https://www.rsna.org/Machine-Learning-Showcase/
  90. 90.
  91. 91.
    Jacobson I. Object-oriented development in an industrial environment. ACM SIGPLAN Not. 1987 Dec 1;22 (12):183–191). ACM.Google Scholar
  92. 92.
    Alistair C. Writing effective use cases. Michigan: Addison-Wesley; 2001.Google Scholar
  93. 93.
  94. 94.
    Competitions Kaggle Data Science Bowl. https://www.kaggle.com/c/data-science-bowl-2017
  95. 95.
    Competitions Kaggle Lung Cancer Risk. https://www.kaggle.com/c/msk-redefining-cancer-treatme nt
  96. 96.
  97. 97.
    Competitions Kaggle Seizure Prediction. https://www.kaggle.com/c/seizure-prediction
  98. 98.
    Personal communication. (soon in press_Andriole, Katherine. MGH and BWI Center For Clinical Data Science.Google Scholar
  99. 99.
  100. 100.
  101. 101.
    Ching T, Himmelstein DS, Beaulieu-Jones BK, Kalinin AA, Do BT, Way GP, Ferrero E, Agapow PM, Zietz M, Hoffman MM, Xie W. Opportunities and obstacles for deep learning in biology and medicine. bioRxiv. 2018 Jan;1:142760.Google Scholar
  102. 102.
    Balthazar P, Harri P, Prater A, Safdar NM. Protecting your patients’ interests in the era of big data, artificial intelligence, and predictive analytics. J Am Coll Radiol. 2018 Mar 1;15(3):580–6.CrossRefGoogle Scholar
  103. 103.
    Berinato S. With big data comes big responsibility. Harv Bus Rev. 2014;92(11):20.Google Scholar
  104. 104.
    Merkle RC. A digital signature based on a conventional encryption function. In Conference on the theory and application of cryptographic techniques 1987 Aug 16 (pp. 369–378). Berlin, Heidelberg: Springer.Google Scholar
  105. 105.
    Lazer D, Kennedy R, King G, Vespignani A. The parable of Google Flu: traps in big data analysis. Science. 2014 Mar 14;343(6176):1203–5.CrossRefGoogle Scholar
  106. 106.
  107. 107.
    Ekblaw A, Azaria A, Halamka JD, Lippman A. A case study for blockchain in healthcare: “MedRec” prototype for electronic health records and medical research data. In Proceedings of IEEE Open & Big Data Conference 2016 Aug 22 (vol. 13, p. 13).Google Scholar
  108. 108.
  109. 109.
  110. 110.
  111. 111.
    ACR Data Science Institute Data Science Summit. https://www.acrdsi.org/dsisummit2018
  112. 112.
  113. 113.
  114. 114.
  115. 115.
  116. 116.
    MQSA public Law. PL 102-539.Google Scholar
  117. 117.

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Bibb Allen
    • 1
    • 2
  • Robert Gish
    • 3
  • Keith Dreyer
    • 2
    • 4
  1. 1.Department of RadiologyGrandview Medical CenterBirminghamUSA
  2. 2.American College of Radiology Data Science InstituteRestonUSA
  3. 3.Diagnostic RadiologyBrookwood Baptist HealthBirminghamUSA
  4. 4.Department of RadiologyMassachusetts General HospitalBostonUSA

Personalised recommendations