Abstract
A comprehensive assessment of the impact of eXplainable AI (XAI) on diagnostic decision-making should adopt a socio-technical perspective. Our study focuses on Decision Support Systems (DSS) that provide explanations in the form of Activation Maps, assessing their impact in terms of automation bias and algorithmic aversion. Specifically, we focus on the XAI-assisted task of detecting thoraco-lumbar fractures from X-rays by radiologists, taking into account the complexity of the cases and the experience level of users. Our results show how XAI support has a clear and positive impact on diagnostic performance. By introducing the concepts of technology impact, reliance patterns, and the white box paradox, we highlight the importance of designing Human-AI Collaboration Protocols (HAI-CP) that are specific to the task at hand to optimize the integration of XAI into diagnostic decision-making.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
- 2.
- 3.
The detailed analysis of each kind of XAI support is the subject of another work currently in preparation. In this paper we report on the longitudinal analysis of the impact of XAI without further distinguishing the type of output.
- 4.
Both diagrams were generated via the tool available at https://haiiassassessment.pythonanywhere.com/.
- 5.
Automation bias was calculated via the tool “DSS Quality Assessment" available at https://dss-quality-assessment.vercel.app/?step=4 following the metrics reported in [13].
References
Liew, C.: The future of radiology augmented with artificial intelligence: a strategy for success. Eur. J. Radiol. 102, 152–156 (2018)
Sogani, J., Allen Jr., B., Dreyer, K., McGinty, G.: Artificial intelligence in radiology: the ecosystem essential to improving patient care (2020)
Allen, B., Agarwal, S., Kalpathy-Cramer, J., Dreyer, K.: Democratizing AI. J. Am. Coll. Radiol. 16(7), 961–963 (2019)
Quinn, T.P., Jacobs, S., Senadeera, M., Le, V., Coghlan, S.: The three ghosts of medical AI: can the black-box present deliver? Artif. Intell. Med. 124, 102158 (2022)
Cabitza, F., et al.: Quod erat demonstrandum?-towards a typology of the concept of explanation for the design of explainable AI. Expert Syst. Appl. 213, 118888 (2023)
Cabitza, F., et al.: Rams, hounds and white boxes: investigating human-AI collaboration protocols in medical diagnosis. Artif. Intell. Med. 138, 102506 (2023)
Cabitza, F., Campagner, A., Natali, C., Parimbelli, E., Ronzio, L., Cameli, M.: Painting the black box white: experimental findings from applying XAI to an ECG reading setting. Mach. Learn. Knowl. Extr. 5(1), 269–286 (2023)
Bertrand, A., Belloum, R., Eagan, J.R., Maxwell, W.: How cognitive biases affect XAI-assisted decision-making: a systematic review. In: Proceedings of the 2022 AAAI/ACM Conference on AI, Ethics, and Society, pp. 78–91 (2022)
Mumford, E.: A socio-technical approach to systems design. Requirements Eng. 5, 125–133 (2000)
Delmas, P.D., et al.: Underdiagnosis of vertebral fractures is a worldwide problem: the impact study. J. Bone Mineral Res. 20(4), 557–563 (2005)
Vilone, G., Longo, L.: Classification of explainable artificial intelligence methods through their output formats. Mach. Learn. Knowl. Extract. 3(3), 615–661 (2021)
Skitka, L.J., Mosier, K.L., Burdick, M.: Does automation bias decision-making? Int. J. Hum.-Comput. Stud. 51(5), 991–1006 (1999)
Cabitza, F., Campagner, A., Angius, R., Natali, C., Reverberi, C.: AI shall have no dominion: on how to measure technology dominance in AI-supported human decision-making. In CHI 2023: The Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (2023)
Dietvorst, B.J., Simmons, J., Massey, C.: Understanding algorithm aversion: forecasters erroneously avoid algorithms after seeing them err. In: Academy of Management Proceedings, Briarcliff Manor, NY, USA, no. 1, p. 12227. Academy of Management (2014)
Jussupow, E., Benbasat, I., Heinzl, A.: Why are we averse towards algorithms? A comprehensive literature review on algorithm aversion. In: Twenty-Eighth European Conference on Information Systems (ECIS2020), Atlanta, GA USA, pp. 1–16. AIS (2020)
Sutton, S.G., Arnold, V., Holt, M.: An extension of the theory of technology dominance: understanding the underlying nature, causes and effects. Causes and effects (2022)
Cabitza, F.: Biases affecting human decision making in AI-supported second opinion settings. In: Torra, V., Narukawa, Y., Pasi, G., Viviani, M. (eds.) MDAI 2019. LNCS (LNAI), vol. 11676, pp. 283–294. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-26773-5_25
Cabitza, F., Campagner, A., Famiglini, L., Gallazzi, E., La Maida, G.A.: Color shadows (part I): exploratory usability evaluation of activation maps in radiological machine learning. In: Holzinger, A., Kieseberg, P., Tjoa, A.M., Weippl, E. (eds.) CD-MAKE 2022. LNCS, vol. 13480, pp. 31–50. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-14463-9_3
Zhou, B., Khosla, A., Lapedriza, A., Oliva, A., Torralba, A.: Learning deep features for discriminative localization. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016)
Friedman, C.P.: A “fundamental theorem’’ of biomedical informatics. J. Am. Med. Inform. Assoc. 16(2), 169–170 (2009)
Kamar, E.: Directions in hybrid intelligence: complementing AI systems with human intelligence. In: IJCAI, pp. 4070–4073 (2016)
Gupta, P., Woolley, A.W.: Articulating the role of artificial intelligence in collective intelligence: a transactive systems framework. In: Proceedings of the Human Factors and Ergonomics Society Annual Meeting, Los Angeles, CA, USA, vol. 65, pp. 670–674. SAGE Publications (2021)
Miller, R.A., Masarie Jr., F.E.: The demise of the “Greek oracle” model for medical diagnostic systems. Methods Inf. Med. 29(01), 1–2 (1990)
Campagner, A., Sternini, F., Cabitza, F.: Decisions are not all equal. Introducing a utility metric based on case-wise raters’ perceptions. Comput. Methods Program. Biomed. 106930 (2022)
Dixon-Woods, M., Redwood, S., Leslie, M., Minion, J., Martin, G.P., Coleman, J.J.: Improving quality and safety of care using “technovigilance’’: an ethnographic case study of secondary use of data from an electronic prescribing and decision support system. Milbank Q. 91(3), 424–454 (2013)
Cabitza, F., Zeitoun, J.-D.: The proof of the pudding: in praise of a culture of real-world validation for medical artificial intelligence. Ann. Transl. Med. 7(8) (2019)
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
Natali, C., Famiglini, L., Campagner, A., La Maida, G.A., Gallazzi, E., Cabitza, F. (2023). Color Shadows 2: Assessing the Impact of XAI on Diagnostic Decision-Making. In: Longo, L. (eds) Explainable Artificial Intelligence. xAI 2023. Communications in Computer and Information Science, vol 1901. Springer, Cham. https://doi.org/10.1007/978-3-031-44064-9_33
Download citation
DOI: https://doi.org/10.1007/978-3-031-44064-9_33
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-44063-2
Online ISBN: 978-3-031-44064-9
eBook Packages: Computer ScienceComputer Science (R0)