Abstract
Artificial intelligence (AI) holds tremendous promise for medical image analysis and computer-aided diagnosis, but various challenges must be addressed to enable its widespread adoption and impact in patient care. Open science, specifically through open-source code and public databases, brings multiple benefits to the progress of AI in medical imaging. It is expected to facilitate research output sharing, promote collaboration among researchers, improve the reproducibility of findings, and foster innovation. However, it is important to recognize that the current state of open-source research, particularly with respect to the large, public datasets commonly used in medical imaging AI, is inherently centered around high-income countries (HIC) and privileged populations. Low- and middle-income countries (LMIC) often face several limitations in contributing to and benefiting from open science research in this domain, such as inadequate digital infrastructure, limited funding for research and development, and a scarcity of healthcare and data science professionals. This may lead to further global disparities in health equity as AI-based clinical decision support systems continue to be implemented in practice. While transfer learning and distributed learning hold promise in addressing some challenges related to limited and non-public data in LMIC, practical obstacles arise when dealing with small, lower-quality datasets, resource constraints, and the need for tailored local implementation of these models. In this commentary, we explore the relationship between open-source models and public medical imaging data repositories in the context of transfer learning and distributed learning, specifically considering their implications for global health equity.
R. Souza and E.A.M. Stanley—Contributed equally.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
ADNI | Acquisition Sites. https://adni.loni.usc.edu/about/centers-cores/study-sites/
The Federated Tumor Segmentation (FeTS) initiative. https://www.med.upenn.edu/cbica/fets/#FeTSCollaboratingSites6
Kaggle DR dataset (EyePACS). https://www.kaggle.com/datasets/mariaherrerot/eyepacspreprocess
Ethics and governance of artificial intelligence for health: WHO Guidance (2021). https://www.who.int/publications-detail-redirect/9789240029200
Reporting standards and availability of data, materials, code and protocols (2023). https://www.nature.com/nature-portfolio/editorial-policies/reporting-standards
Annas, G.J.: HIPAA regulations - a new era of medical-record privacy? N. Engl. J. Med. 348(15), 1486–1490 (2003)
Ashford, M.T., et al.: Screening and enrollment of underrepresented ethnocultural and educational populations in the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Alzheimer’s & Dementia 18(12), 2603–2613 (2022)
Beam, A.L., Manrai, A.K., Ghassemi, M.: Challenges to the reproducibility of machine learning models in health care. JAMA 323(4), 305–306 (2020)
Bosma, J., et al.: Reproducibility of Training Deep Learning Models for Medical Image Analysis. Medical Imaging with Deep Learning (2023). https://openreview.net/forum?id=MR01DcGST9
Ciecierski-Holmes, T., Singh, R., Axt, M., Brenner, S., Barteit, S.: Artificial intelligence for strengthening healthcare systems in low- and middle-income countries: a systematic scoping review. NPJ Digit. Med. 5(1), 1–13 (2022)
Coates, M.M., et al.: Burden of disease among the world’s poorest billion people: an expert-informed secondary analysis of global burden of disease estimates. PLoS ONE 16(8), e0253073 (2021)
Diaz, O., et al.: Data preparation for artificial intelligence in medical imaging: a comprehensive guide to open-access platforms and tools. Physica Med. 83, 25–37 (2021)
Fry, A., et al.: Comparison of sociodemographic and health-related characteristics of UK biobank participants with those of the general population. Am. J. Epidemiol. 186(9), 1026–1034 (2017)
Ganguli, M., Lee, C.W., Hughes, T., Snitz, B.E., Jakubcak, J., Duara, R., Chang, C.C.H.: Who wants a free brain scan? Assessing and correcting for recruitment biases in a population-based sMRI pilot study. Brain Imaging Behav. 9(2), 204–212 (2015)
Hyett, S., Marjerrison, S., Gabel, C.: Improving health research among Indigenous Peoples in Canada. CMAJ 190(20), E616–E621 (2018)
Jack, C.R., et al.: The Alzheimer’s disease neuroimaging initiative (ADNI): MRI methods. J Magn. Reson. Imaging 27(4), 685–691 (2008)
Kwet, M.: Digital colonialism: US empire and the new imperialism in the Global South. Race Class 60(4), 3–26 (2019)
Lo Vercio, L., et al.: Supervised machine learning tools: a tutorial for clinicians. J. Neural Eng. 17(6), 062001 (2020)
MacEachern, S.J., Forkert, N.D.: Machine learning for precision medicine. Genome 64(4), 416–425 (2021)
McDermott, M.B.A., Wang, S., Marinsek, N., Ranganath, R., Foschini, L., Ghassemi, M.: Reproducibility in machine learning for health research: still a ways to go. Sci. Transl. Med. 13(586), eabb1655 (2021)
McMahan, B., Moore, E., Ramage, D., Hampson, S., Arcas, B.A.y.: Communication-efficient learning of deep networks from decentralized data. In: Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, pp. 1273–1282. PMLR (2017)
Menze, B.H., et al.: The multimodal brain tumor image segmentation benchmark (BRATS). IEEE Trans. Med. Imaging 34(10), 1993–2024 (2015)
Mollura, D.J., et al.: Artificial Intelligence in low- and middle-income countries: innovating global health radiology. Radiology 297(3), 513–520 (2020)
Ng, D., Lan, X., Yao, M.M.S., Chan, W.P., Feng, M.: Federated learning: a collaborative effort to achieve better medical imaging models for individual sites that have small labelled datasets. Quant. Imaging Med. Surg. 11(2), 852–857 (2021)
Nguyen, N.H., Subhan, F.B., Williams, K., Chan, C.B.: Barriers and mitigating strategies to healthcare access in indigenous communities of Canada: a narrative review. Healthcare 8(2), 112 (2020)
Okolo, C.T.: Optimizing human-centered AI for healthcare in the Global South. Patterns 3(2), 100421 (2022)
Pati, S., et al.: The federated tumor segmentation (FeTS) challenge, May 2021. arXiv:2105.05874 [cs, eess]
Pineau, J., et al.: Improving reproducibility in machine learning research (a report from the NeurIPS 2019 reproducibility program). J. Mach. Learn. Res. 22(1), 164:7459–164:7478 (2021)
Porwal, P.: Indian Diabetic Retinopathy Image Dataset (IDRiD) (2018). https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid
Prevedello, M., et al.: Challenges related to artificial intelligence research in medical imaging and the importance of image analysis competitions. Radiol. Artif. Intell. 1(1), e180031 (2019)
Salman, H., Jain, S., Ilyas, A., Engstrom, L., Wong, E., Madry, A.: When does bias transfer in transfer learning? (2022). arXiv:2207.02842 [cs]
Selbst, A.D., Boyd, D., Friedler, S.A., Venkatasubramanian, S., Vertesi, J.: Fairness and abstraction in sociotechnical systems. In: Proceedings of the Conference on Fairness, Accountability, and Transparency. pp. 59–68. FAT* 2019. Association for Computing Machinery, New York (2019)
Sivaswamy, J., Krishnadas, S.R., Datt Joshi, G., Jain, M., Syed Tabish, A.U.: Drishti-GS: retinal image dataset for optic nerve head(ONH) segmentation. In: 2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI), pp. 53–56 (2014)
Souza, R., Mouches, P., Wilms, M., Tuladhar, A., Langner, S., Forkert, N.D.: An analysis of the effects of limited training data in distributed learning scenarios for brain age prediction. J. Am. Med. Inform. Assoc. 30(1), 112–119 (2023)
Souza, R., et al.: Image-encoded biological and non-biological variables may be used as shortcuts in deep learning models trained on multi-site neuroimaging data. J. Am. Med. Inform. Assoc. (2023). https://doi.org/10.1093/jamia/ocad171
Sudlow, C., et al.: UK Biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 12(3), e1001779 (2015)
Tuladhar, A., Rajashekar, D., Forkert, N.D.: Distributed learning in healthcare. In: Sakly, H., Yeom, K., Halabi, S., Said, M., Seekins, J., Tagina, M. (eds.) Trends of Artificial Intelligence and Big Data for E-Health, pp. 183–212. Springer, Cham, Integrated Science (2022). https://doi.org/10.1007/978-3-031-11199-0_10
Wang, K.: Opportunities in open science with AI. Front. Big Data 2, 26 (2019)
Weiss, K., Khoshgoftaar, T.M., Wang, D.: A survey of transfer learning. J. Big Data 3(1), 9 (2016)
Williams, D., Hornung, H., Nadimpalli, A., Peery, A.: Deep learning and its application for healthcare delivery in low and middle income countries. Front. Artif. Intelli. 4, 553987 (2021)
Zhou, Y., Elashoff, D., Kremen, S., Teng, E., Karlawish, J., Grill, J.D.: African Americans are less likely to enroll in preclinical Alzheimer’s disease clinical trials. Alzheimer’s Dementia Transl. Res. Clin. Intervent. 3(1), 57–64 (2017)
Zuboff, S.: Big other: surveillance capitalism and the prospects of an information civilization. J. Inf. Technol. 30(1), 75–89 (2015)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Souza, R., Stanley, E.A.M., Forkert, N.D. (2023). On the Relationship Between Open Science in Artificial Intelligence for Medical Imaging and Global Health Equity. In: Wesarg, S., et al. Clinical Image-Based Procedures, Fairness of AI in Medical Imaging, and Ethical and Philosophical Issues in Medical Imaging. CLIP EPIMI FAIMI 2023 2023 2023. Lecture Notes in Computer Science, vol 14242. Springer, Cham. https://doi.org/10.1007/978-3-031-45249-9_28
Download citation
DOI: https://doi.org/10.1007/978-3-031-45249-9_28
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-45248-2
Online ISBN: 978-3-031-45249-9
eBook Packages: Computer ScienceComputer Science (R0)