BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20250327T203344Z
LOCATION:Frontenac Foyer
DTSTART;TZID=America/New_York:20250331T164500
DTEND;TZID=America/New_York:20250331T181500
UID:HFESHCS_2025 International Symposium on Human Factors and Ergonomics i
 n Health Care_sess147_POST185@linklings.com
SUMMARY:DH6 - Bridging the Gap: The Impact of Multimodal AI on Clinician T
 rust and Acceptance in Ophthalmic Care
DESCRIPTION:Poster Presentation\n\nFaisal Ghaffar, Yousuf Zia Islam, Nadin
 e Furtado, and Catherine Burns (University of Waterloo)\n\nIn this study, 
 we explored the integration of artificial intelligence (AI) in healthcare,
  specifically in ophthalmology, with a focus on the differences between un
 imodal and multimodal AI systems and their impact on clinician trust and u
 ser acceptance. AI has shown potential in the diagnosis and management of 
 complex eye diseases such as glaucoma, a leading cause of irreversible bli
 ndness globally (Sun et al., 2022). These AI systems rely on either a sing
 le type of diagnostic data, termed as unimodal (Fu et al., 2018), or combi
 ne multiple data sources, such as fundus imaging, visual field tests, and 
 patient histories, which is called a multimodal system (Lim et al., 2022).
  These AI systems aim to improve diagnostic accuracy and support clinical 
 decision-making. However, the effectiveness of these systems is not determ
 ined solely by their technical accuracy, but also by how well they align w
 ith human mental models and how much trust they inspire in the users.\nTru
 st and user acceptance are critical factors in the successful adoption of 
 AI systems, especially in healthcare where errors can have severe conseque
 nces. Trust in AI systems is influenced by factors such as input data, inf
 ormation and system quality, accuracy and transparency (Cai et al., 2019; 
 Lukyanenko et al., 2022; Miller, 2019). It is also necessary to keep an ap
 propriate balance of trust; insufficient trust leads to underutilization, 
 while over-reliance can result in misuse or overconfidence in AI recommend
 ations (Parasuraman & Riley, 1997). User acceptance refers to the degree t
 o which users are willing to adopt and use a new system.\nWe developed and
  evaluated both unimodal and multimodal AI mockup systems designed for gla
 ucoma diagnosis and interviewed 20 optometrists to assess their trust in e
 ach system across five key dimensions: input data modality, information qu
 ality, technical accuracy, output explainability, and system quality. We f
 urther used a modified version of a unified theory of acceptance and use o
 f technology model to evaluate user acceptance and predict the likelihood 
 of successful adoption of unimodal and multimodal systems by examining fac
 tors such as modality preference, performance expectancy, effort expectanc
 y, social influence, and facilitating conditions,\nThe results showed a cl
 ear preference for multimodal systems, which received an average trust rat
 ing of 2.0 compared to 2.9 for unimodal systems, with a statistically sign
 ificant difference (p < .001) on the Wilcoxon signed-rank test. Across all
  five dimensions, the multimodal system consistently outperformed the unim
 odal system, demonstrating that optometrists trust the multimodal system m
 ore than the unimodal system. Similarly, the user acceptance prefers a mul
 timodal system with an average rating of 1.8 compared to 2.5 with a statis
 tically significant difference of P<0.001 on the Wilcoxon signed-rank test
 . Across all dimensions of the UTAUT model, the multimodal has higher acce
 ptance than the unimodal, which shows that users would accept the multimod
 al system more than the unimodal system for glaucoma diagnosis. \nIn terms
  of diagnostic performance, optometrists achieved higher accuracy metrics 
 with the multimodal system. The multimodal AI system demonstrated superior
  precision (71.43%), recall (88.71%), and F1 score (79.14%), while the uni
 modal system produced lower results, with precision at 60%, recall at 67.9
 2%, and F1 score at 63.72%. These findings suggest that the increased trus
 t in multimodal systems directly correlates with improved diagnostic outco
 mes, making these systems more effective in clinical practice.\n\nKey take
 away points:\n▪Aligning AI systems with clinicians' mental models and work
 flows is critical for successful collaboration and decision-making in heal
 thcare environments.\n▪ Trust in AI systems is essential for their effecti
 ve implementation in healthcare. Multimodal AI systems inspire greater use
 r trust and have higher user acceptance due to their use of diverse data s
 ources and their better alignment with clinicians' decision-making process
 es.\n▪The study shows a statistically significant higher trust and higher 
 acceptance in multimodal systems compared to unimodal systems), and this t
 rust and user acceptance is strongly associated with better diagnostic per
 formance.\n▪The multimodal system had higher diagnostic accuracy, precisio
 n, recall, and F1 score compared to the unimodal system, indicating that c
 ombining multiple data modalities improves not only trust but also clinica
 l outcomes.\nThese findings provide valuable insights for the design and d
 eployment of AI systems in healthcare, especially for conditions like glau
 coma, where precision is crucial for preventing irreversible vision loss. 
 The results emphasize the importance of considering trust and system desig
 n when implementing AI technologies in clinical practice.\n\nReferences\nC
 ai, C. J., Winter, S., Steiner, D., Wilcox, L., & Terry, M. (2019). “Hello
  AI”: Uncovering the Onboarding Needs of Medical Practitioners for Human-A
 I Collaborative Decision-Making. Proceedings of the ACM on Human-Computer 
 Interaction, 3(CSCW), 1–24. https://doi.org/10.1145/3359206\nFu, H., Cheng
 , J., Xu, Y., Zhang, C., Wong, D. W. K., Liu, J., & Cao, X. (2018). Disc-A
 ware Ensemble Network for Glaucoma Screening From Fundus Image. IEEE Trans
 actions on Medical Imaging, 37(11), 2493–2501. https://doi.org/10.1109/TMI
 .2018.2837012\nLim, W. S., Ho, H.-Y., Ho, H.-C., Chen, Y.-W., Lee, C.-K., 
 Chen, P.-J., Lai, F., Jang, J.-S. R., & Ko, M.-L. (2022). Use of multimoda
 l dataset in AI for detecting glaucoma based on fundus photographs assesse
 d with OCT: focus group study on high prevalence of myopia. BMC Medical Im
 aging, 22(1), 206. https://doi.org/10.1186/s12880-022-00933-z\nLukyanenko,
  R., Maass, W., & Storey, V. C. (2022). Trust in artificial intelligence: 
 From a Foundational Trust Framework to emerging research opportunities. El
 ectronic Markets, 32(4), 1993–2020. https://doi.org/10.1007/s12525-022-006
 05-4\nMiller, T. (2019). Explanation in artificial intelligence: Insights 
 from the social sciences. Artificial Intelligence, 267, 1–38. https://doi.
 org/10.1016/j.artint.2018.07.007\nParasuraman, R., & Riley, V. (1997). Hum
 ans and Automation: Use, Misuse, Disuse, Abuse. Human Factors: The Journal
  of the Human Factors and Ergonomics Society, 39(2), 230–253. https://doi.
 org/10.1518/001872097778543886\nSun, Y., Chen, A., Zou, M., Zhang, Y., Jin
 , L., Li, Y., Zheng, D., Jin, G., & Congdon, N. (2022). Time trends, assoc
 iations and prevalence of blindness and vision loss due to glaucoma: an an
 alysis of observational data from the Global Burden of Disease Study 2017.
  BMJ Open, 12(1), e053805. https://doi.org/10.1136/bmjopen-2021-053805\n\n
 Track: Digital Health (DH), Simulation and Education (SE), Hospital Enviro
 nments (HE), Medical and Drug Delivery Devices (MDD), Patient Safety and R
 esearch Initiatives (PS)
END:VEVENT
END:VCALENDAR