Risks and benefits of dermatological machine learning health care applications-an overview and ethical analysis.
Journal
Journal of the European Academy of Dermatology and Venereology : JEADV
ISSN: 1468-3083
Titre abrégé: J Eur Acad Dermatol Venereol
Pays: England
ID NLM: 9216037
Informations de publication
Date de publication:
Sep 2022
Sep 2022
Historique:
received:
05
10
2021
accepted:
07
04
2022
pubmed:
2
5
2022
medline:
23
8
2022
entrez:
1
5
2022
Statut:
ppublish
Résumé
Visual data are particularly amenable for machine learning techniques. With clinical photography established for skin surveillance and documentation purposes as well as progress checks, dermatology is an ideal field for the development and application of emerging machine learning health care applications (ML-HCAs). To date, several ML-HCAs have detected malignant skin lesions on par with experts or found overlooked visual patterns that correlate with certain dermatological diseases. However, it is well established that ML-HCAs come with ethical and social implications. Currently, there is a lack of research that establishes model design, training, usage and regulation of such technologies sufficient to ensure ethically and socially responsible development and clinical translation, specifically within the field of dermatology. With this paper, we aim to give an overview of currently discussed ethical issues relating to dermatological ML-HCAs. On the basis of a thematic, keyword-based literature search, we performed an ethical analysis against established frameworks of biomedical ethics. We combined our results with current, relevant normative machine learning ethics literature to identify the status quo of the ethics of ML-HCAs in dermatology. We describe the benefits and risks of dermatological ML-HCAs that are currently being developed for clinical purposes. The potential benefits range from better patient outcomes to better knowledge accessibility to decreasing health care disparities, that is, standards of care between different population groups. The risks associated with ML-HCAs range from confidentiality issues to individual patient outcomes as well as the exacerbation of prevalent health care disparities. We discuss the practical implications for all stages of dermatological ML-HCA development. We found that ML-HCAs present stakeholder-specific risks for patients, health care professionals and society, which need to be considered separately. The discipline lacks sufficient biomedical ethics research that could standardize the approach to ML-HCA model design, training, use and regulation of such technologies.
Sections du résumé
BACKGROUND
BACKGROUND
Visual data are particularly amenable for machine learning techniques. With clinical photography established for skin surveillance and documentation purposes as well as progress checks, dermatology is an ideal field for the development and application of emerging machine learning health care applications (ML-HCAs). To date, several ML-HCAs have detected malignant skin lesions on par with experts or found overlooked visual patterns that correlate with certain dermatological diseases. However, it is well established that ML-HCAs come with ethical and social implications.
OBJECTIVES
OBJECTIVE
Currently, there is a lack of research that establishes model design, training, usage and regulation of such technologies sufficient to ensure ethically and socially responsible development and clinical translation, specifically within the field of dermatology. With this paper, we aim to give an overview of currently discussed ethical issues relating to dermatological ML-HCAs.
METHODS
METHODS
On the basis of a thematic, keyword-based literature search, we performed an ethical analysis against established frameworks of biomedical ethics. We combined our results with current, relevant normative machine learning ethics literature to identify the status quo of the ethics of ML-HCAs in dermatology. We describe the benefits and risks of dermatological ML-HCAs that are currently being developed for clinical purposes.
RESULTS
RESULTS
The potential benefits range from better patient outcomes to better knowledge accessibility to decreasing health care disparities, that is, standards of care between different population groups. The risks associated with ML-HCAs range from confidentiality issues to individual patient outcomes as well as the exacerbation of prevalent health care disparities. We discuss the practical implications for all stages of dermatological ML-HCA development.
CONCLUSION
CONCLUSIONS
We found that ML-HCAs present stakeholder-specific risks for patients, health care professionals and society, which need to be considered separately. The discipline lacks sufficient biomedical ethics research that could standardize the approach to ML-HCA model design, training, use and regulation of such technologies.
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
1660-1668Subventions
Organisme : Bundesministerium für Gesundheit
ID : 2520DAT920
Informations de copyright
© 2022 The Authors. Journal of the European Academy of Dermatology and Venereology published by John Wiley & Sons Ltd on behalf of European Academy of Dermatology and Venereology.
Références
LeCun Y, Bengio Y, Hinton G. Deep learning. Nature 2015; 521: 436-444.
Adamson AS, Smith A. Machine learning and health care disparities in dermatology. JAMA Dermatol 2018; 154: 1247-1248.
Kunde L, McMeniman E, Parker M. Clinical photography in dermatology: ethical and medico-legal considerations in the age of digital and smartphone technology. Australas J Dermatol 2013; 54: 192-197.
Esteva A, Kuprel B, Novoa RA et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature 2017; 542: 115-118.
Tschandl P, Rinner C, Apalla Z et al. Human-computer collaboration for skin cancer recognition. Nat Med 2020; 26: 1229-1234. https://www.nature.com/articles/s41591-020-0942-0
Tongdee E, Markowitz O. Mobile app rankings in dermatology. Cutis 2018; 102: 252-256. https://pubmed.ncbi.nlm.nih.gov/30489561/
Flaten HK, St Claire C, Schlager E, Dunnick CA, Dellavalle RP. Growth of mobile applications in dermatology - 2017 update. Dermatol Online J 2018; 24. doi:10.5070/D3242038180
Cho SI, Sun S, Mun J-H et al. Dermatologist-level classification of malignant lip diseases using a deep convolutional neural network. Br J Dermatol 2020; 182: 1388-1394.
Han SS, Moon IJ, Lim W et al. Keratinocytic skin cancer detection on the face using region-based convolutional neural network. JAMA Dermatol 2020; 156: 29-37.
Han SS, Park GH, Lim W et al. Deep neural networks show an equivalent and often superior performance to dermatologists in onychomycosis diagnosis: automatic construction of onychomycosis datasets by region-based convolutional deep neural network. PLoS One 2018; 13: e0191493.
Dopytalska K, Sobolewski P, Błaszczak A, Szymańska E, Walecka I. Psoriasis in special localizations. Reumatologia 2018; 56: 392-398.
Floridi L, Cowls J, Beltrametti M et al. AI4People-an ethical framework for a good AI society: opportunities, risks, principles, and recommendations. Minds Mach 2018; 28: 689-707.
Mittelstadt B. Principles alone cannot guarantee ethical AI. Nat Mach Intell 2019; 1: 501-507.
Fiske A, Tigard D, Müller R, Haddadin S, Buyx A, McLennan S. Embedded ethics could help implement the pipeline model framework for machine learning healthcare applications. Am J Bioeth 2020; 20: 32-35.
Vayena E, Blasimme A, Cohen IG. Machine learning in medicine: addressing ethical challenges. PLOS medicine 2018; 15: e1002689. https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002689
Jobin A, Ienca M, Vayena E. The global landscape of AI ethics guidelines. Nat Mach Intell 2019; 1: 389-399.
Char DS, Shah NH, Magnus D. Implementing machine learning in health care - addressing ethical challenges. N Engl J Med 2018; 378: 981-983.
Thomsen K, Iversen L, Titlestad TL, Winther O. Systematic review of machine learning for diagnosis and prognosis in dermatology. J Dermatolog Treat 2020; 31: 496-510.
Eapen BR. Artificial intelligence in dermatology: a practical introduction to a paradigm shift. Indian Dermatol Online J 2020; 11: 881-889.
Chan S, Reddy V, Myers B, Thibodeaux Q, Brownstone N, Liao W. Machine learning in dermatology: current applications, opportunities, and limitations. Dermatol Ther 2020; 10: 365-386.
De A, Sarda A, Gupta S, Das S. Use of artificial intelligence in dermatology. Indian J Dermatol 2020; 65: 352-357.
Brinker TJ, Schmitt M, Krieghoff-Henning EI et al. Diagnostic performance of artificial intelligence for histologic melanoma recognition compared to 18 international expert pathologists. J Am Acad Dermatol 2021; 86: 640-642.
Brinker TJ, Hekler A, Enk AH et al. A convolutional neural network trained with dermoscopic images performed on par with 145 dermatologists in a clinical melanoma image classification task. Eur J Cancer 2019; 111: 148-154.
Fujisawa Y, Otomo Y, Ogata Y et al. Deep-learning-based, computer-aided classifier developed with a small dataset of clinical images surpasses board-certified dermatologists in skin tumour diagnosis. Br J Dermatol 2019; 180: 373-381.
Haenssle HA, Fink C, Schneiderbauer R et al. Man against machine: diagnostic performance of a deep learning convolutional neural network for dermoscopic melanoma recognition in comparison to 58 dermatologists. Ann Oncol 2018; 29: 1836-1842.
Brinker T, Hekler A, Enk A, Klode J, Hauschild A, Berking C et al. Deep learning outperformed 136 of 157 dermatologists in a head-to-head dermoscopic melanoma image classification task. Eur J Cancer 2019; 113:47-54. URL https://www.ejcancer.com/article/S0959-8049(19)30221-7/fulltext
Barnett ML, Song Z, Landon BE. Trends in physician referrals in the United States, 1999-2009. Arch Intern Med 2012; 172: 163-170. https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/1108675
Farberg AS, Winkelmann RR, Tucker N, White R, Rigel DS. The impact of quantitative data provided by a multi-spectral digital skin lesion analysis device on Dermatologists'Decisions to biopsy pigmented lesions. J Clin Aesthet Dermatol 2017; 10: 24-26.
Topol EJ, Verghese A. Deep Medicine: How Artificial Intelligence can make Healthcare Human Again, 1st edn. Hachette, UK, 2019.
Toth J, Bartha L, Szabo T, Lazar I, Harangi B, Hajdu A. An online application for storing, analyzing, and sharing dermatological data. 2015 6th IEEE International Conference on Cognitive Infocommunications (CogInfoCom). IEEE, 2015: 339-342.
Gomolin A, Netchiporouk E, Gniadecki R, Litvinov IV. Artificial intelligence applications in dermatology: where do we stand? Front Med 2020; 7: 100.
Zakhem GA, Motosko CC, Ho RS. How should artificial intelligence screen for skin cancer and deliver diagnostic predictions to patients? JAMA Dermatol 2018; 154: 1383-1384. https://pubmed.ncbi.nlm.nih.gov/30140854/
Madabhushi A, Lee G. Image analysis and machine learning in digital pathology: challenges and opportunities. Med Image Anal 2016; 33: 170-175.
Wongvibulsin S, Ho BK-T, Kwatra SG. Embracing machine learning and digital health technology for precision dermatology. J Dermatolog Treat 2020; 31: 494-495.
Yu K-H, Beam AL, Kohane IS. Artificial intelligence in healthcare. Nat Biomed Eng 2018; 2: 719-731. https://www.nature.com/articles/s41551-018-0305-z
World Health Organization, OECD, and International Bank for Reconstruction and Development/The World Bank. Delivering quality health services. A global imperative for universal health coverage. 2018 [cited 2021 Jun 7]. URL https://apps.who.int/iris/bitstream/handle/10665/272465/9789241513906-eng.pdf?ua=1
Arimany Manso J, Taberner Ferrer R, Pidevall I, Mascaró Ballester JM, Martin-Fumadó C. Use of photography in dermatology: ethical and legal implications. Actas Dermosifiliogr 2020; 111: 107-114.
Binol H, Plotner A, Sopkovich J, Kaffenberger B, Niazi MKK, Gurcan MN. Ros-NET: a deep convolutional neural network for automatic identification of rosacea lesions. Skin Res Technol 2020; 26: 413-421.
Navarrete-Dechent, C, Dusza, S W., Liopyris, K, Marghoob, A A, Halpern, A C, & Marchetti, M A. Automated dermatological diagnosis: hype or reality? 2018. URL https://www.sci-hub.do/https://www.ncbi.nlm.nih.gov/pmc/articles/pmc7701995/
Arimany Manso J, Taberner Ferrer R, Pidevall I, Mascaró Ballester JM, Martin-Fumadó C. Implicaciones bioéticas y médico-legales del uso de la fotografía en dermatología. Actas Dermosifiliogr 2020; 111: 107-114.
Bibbins-Domingo K, Grossman DC, Curry SJ et al. Screening for skin cancer: US preventive services task force recommendation statement. JAMA 2016; 316: 429-435.
Cramb SM, Moraga P, Mengersen KL, Baade PD. Spatial variation in cancer incidence and survival over time across Queensland. Aust Spat Spatio-temporal Epidemiol 2017; 23: 59-67.
Parikh PB, Yang J, Leigh S et al. The impact of financial barriers on access to care, quality of care and vascular morbidity among patients with diabetes and coronary heart disease. J Gen Intern Med 2014; 29: 76-81.
Cioms. International Ethical Guidelines for Biomedical Research Involving Human Subjects. CIOMS, Geneva, 2002.
Pot M, Kieusseyan N, Prainsack B. Not all biases are bad: equitable and inequitable biases in machine learning and radiology. Insights Imag 2021; 12: 13. URL https://insightsimaging.springeropen.com/articles/10.1186/s13244-020-00955-7
Strubell E, Ganesh A, McCallum A. Energy and Policy Considerations for Deep Learning in NLP; 2019 Jun 5. URL https://arxiv.org/pdf/1906.02243
Bender EM, Gebru T, McMillan-Major A, Shmitchell S. On the dangers of stochastic parrots. In: Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency. ACM, New York, NY, 2021: 610-23 [cited 2021 Mar 8].
Polesie S, Gillstedt M, Kittler H et al. Attitudes towards artificial intelligence within dermatology: an international online survey. Br J Dermatol 2020; 183: 159-161.
Blease C, Kaptchuk TJ, Bernstein MH, Mandl KD, Halamka JD, DesRoches CM. Artificial intelligence and the future of primary care: exploratory qualitative study of UKgeneral Practitioners' views. J Med Internet Res 2019; 21: e12802.
Vellido A. Societal issues concerning the application of artificial intelligence in medicine. Kidney Dis 2019; 5: 11-17.
Cormier JN, Xing Y, Ding M et al. Ethnic differences among patients with cutaneous melanoma. Arch Intern Med 2006; 166: 1907-1914.
Ward-Peterson M, Acuña JM, Alkhalifah MK et al. Association between race/ethnicity and survival of melanoma patients in the United States over 3 decades: a secondary analysis of SEER data. Medicine 2016; 95: e3315.
Hendrycks D, Gimpel K. A baseline for detecting misclassified and out-of-distribution examples in neural networks. International conference on learning representations. URL https://arxiv.org/pdf/1610.02136
Wachter S, Mittelstadt B, Russell C. Bias Preservation in Machine Learning: The Legality of Fairness Metrics Under EU Non-Discrimination Law; 2021.
Gonen H, Goldberg Y. Lipstick on a Pig: Debiasing Methods Cover up Systematic Gender Biases in Word Embeddings But do not Remove Them; 2019 Mar 9 [cited 2021 Jan 14]. URL https://arxiv.org/pdf/1903.03862.pdf
Han SS, Kim MS, Lim W, Park GH, Park I, Chang SE. Classification of the clinical images for benign and malignant cutaneous tumors using a deep learning algorithm. J Invest Dermatol 2018; 138: 1529-1538. https://pubmed.ncbi.nlm.nih.gov/29428356/
Young AT, Xiong M, Pfau J, Keiser MJ, Wei ML. Artificial intelligence in dermatology: a primer. J Invest Dermatol 2020; 140: 1504-1512.
Dhar P. The carbon impact of artificial intelligence. Nat Mach Intell 2020; 2: 423-425.
Char DS, Abràmoff MD, Feudtner C. Identifying ethical considerations for machine learning healthcare applications. Am J Bioeth 2020; 20: 7-17.
Obermeyer Z, Emanuel EJ. Predicting the future - big data, machine learning, and clinical medicine. N Engl J Med 2016; 375: 1216-1219.
McLennan S, Lee MM, Fiske A, Celi LA. AI ethics is not a panacea. Am J Bioeth 2020; 20: 20-22.
Hosseinzadeh Kassani S, Hosseinzadeh Kassani P. A comparative study of deep learning architectures on melanoma detection. Tissue Cell 2019; 58: 76-83.
Kinyanjui NM, Odonga T, Cintas C, Codella NCF, Panda R, Sattigeri P, et al. Fairness of Classifiers Across Skin Tones in Dermatology. Springer, Cham; 2020. p. 320-9. URL https://link.springer.com/chapter/10.1007/978-3-030-59725-2_31
Greene D, Hoffmann AL, Stark L, Better, Nicer, Clearer, Fairer: A Critical Assessment of the Movement for Ethical Artificial Intelligence and Machine Learning; 2019. URL https://scholarspace.manoa.hawaii.edu/handle/10125/59651
Rajkomar A, Hardt M, Howell MD, Corrado G, Chin MH. Ensuring fairness in machine learning to advance health equity. Ann Intern Med 2018; 169: 866-872.
Selbst AD, Boyd D, Friedler SA, Venkatasubramanian S, Vertesi J. Fairness and Abstraction in Sociotechnical Systems 2019:59-68.
McLennan S, Fiske A, Celi LA et al. An embedded ethics approach for AI development. Nat Mach Intell 2020; 2: 488-490. https://www.nature.com/articles/s42256-020-0214-1.epdf?sharing_token=gASpBSiQrcnNm6NE6JHEb9RgN0jAjWel9jnR3ZoTv0NG1A26i6bdo3-N5eU1v5o7UE83LU9eU2nf_iiYTjoohG_UktBjTwI3CavZxlc07zurH98PNL41FKS8BKAn47daSxkLvhWmhdyj3V6o3jCqDrmLDuRI0nlpOfeDnJ6NVYk%3D
Bird S, Hutchinson B, Kenthapadi K, Kıcıman E, Mitchell M. Fairness-aware machine learning: practical challenges and lessons learned. In: Companion Proceedings of The 2019 World Wide Web Conference. New York, NY: Association for Computing Machinery; 2019. p. 1297-8 (ACM Digital Library).
Gordon L, Grantcharov T, Rudzicz F. Explainable artificial intelligence for safe intraoperative decision support. JAMA Surg 2019; 154: 1064-1065.
Holzinger A. From Machine Learning to Explainable AI. 2018 World Symposium on Digital Intelligence for Systems and Machines (DISA). [Place of publication not identified]. IEEE, London, UK, 2018: 55-66.
Ghassemi M, Naumann T, Schulam P, Beam AL, Chen IY, Ranganath R. A review of challenges and opportunities in machine learning for health. AMIA Jt Summits Transl Sci Proc 2020; 2020: 191-200.
Ribeiro M, Singh S, Guestrin C. “Why Should I Trust You?”: Explaining the Predictions of Any Classifier. Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations. Association for Computational Linguistics, Stroudsburg, PA, 2016.
Luengo-Oroz M. Solidarity should be a core ethical principle of AI. Nat Mach Intell 2019; 1: 494. URL https://www.nature.com/articles/s42256-019-0115-3
Berle I. Clinical photography and patient rights: the need for orthopraxy. J Med Ethics 2008; 34: 89-92.
GDPR.eu. General Data Protection Regulation (GDPR); 2021 [cited 2021 Apr 29]. URL https://gdpr.eu/
ASPE. Health Insurance Portability and Accountability Act of 1996; 2015 [cited 2021 Apr 29]. URL https://aspe.hhs.gov/report/health-insurance-portability-and-accountability-act-1996
World Medical Association. Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects; 2021 [cited 2021 Apr 21]. URL https://www.wma.net/policies-post/wma-declaration-of-helsinki-ethical-principles-for-medical-research-involving-human-subjects/
Meskó B, Görög M. A short guide for medical professionals in the era of artificial intelligence. NPJ Digit Med. 2020; 3:126. URL https://www.nature.com/articles/s41746-020-00333-z
Obermeyer Z, Lee TH. Lost in thought - the limits of the human mind and the future of medicine. N Engl J Med 2017; 377: 1209-1211.
Manrai AK, Bhatia G, Strymish J, Kohane IS, Jain SH. Medicine's uncomfortable relationship with math: calculating positive predictive value. JAMA Intern Med 2014; 174: 991-993.
Cai CJ, Reif E, Hegde N, Hipp J, Kim B, Smilkov D et al. Human-centered tools for coping with imperfect algorithms during medical decision-making. In: CHI 2019: Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, May 4-9, 2019, Glasgow, Scotland UK. New York, NY: ACM; 2019. p. 1-14.
Cohen IG. Informed Consent and Medical Artificial Intelligence: What to Tell the Patient? SSRN Journal 2020.
DICOM. DICOM; 2021 [cited 2021 Jan 18]. URL https://www.dicomstandard.org/
ISDIS. ISIC Project; 2021 [cited 2021 May 5]. URL https://isdis.org/isic-project/
FDA. Artificial Intelligence and Machine Learning in Software as a Medical Device. FDA 2021 Jan 11 [cited 2021 Apr 28]. URL https://www.fda.gov/medical-devices/software-medical-device-samd/artificial-intelligence-and-machine-learning-software-medical-device