pith. sign in

arxiv: 2607.00542 · v1 · pith:HQMHWPASnew · submitted 2026-07-01 · 💻 cs.HC

AI-Centered Grand Challenges in Visual Analytics for Healthcare: Synthesizing the VAHC 2025 Community Experience

J\"urgen Bernard , David Gotz , Robert S Laramee , Silvia Miksch , Gabriela Morgenshtern , Renata G. Raidou , Alessio Arleo This is my paper

Pith reviewed 2026-07-02 06:28 UTC · model grok-4.3

classification 💻 cs.HC
keywords visual analyticshealthcareAI challengesgrand challengestrust and biasexplainabilityhuman-AI interactionmodel validation
0
0 comments X

The pith

A 2025 workshop synthesis identifies five AI-centered grand challenges for visual analytics in healthcare.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper draws on thematic coding of 15 accepted papers and structured discussions among more than 40 participants at the VAHC 2025 workshop to surface the dominant AI-related issues facing the field. These issues are organized into five clusters that each receive recommendations for future work crossing visualization, AI, and healthcare boundaries. A sympathetic reader would care because the clusters point to concrete places where progress could make AI-driven visual tools more usable and trustworthy in clinical settings. The work frames the challenges as a shared starting point for collaboration rather than isolated technical fixes.

Core claim

Across the three discussion groups, AI surfaced as the most consistent concern. The authors contextualize the workshop outputs against existing literature under five grand-challenge themes and distill the material into five challenge clusters—trust and bias, data and infrastructure, explainability and communication, human-AI interaction, and model reliability and validation—each accompanied by targeted research-direction recommendations.

What carries the argument

Thematic coding of the 15 workshop papers plus structured group discussions organized around three themes, followed by post-workshop reflexive analysis that produces the five challenge clusters.

If this is right

  • Work on trust and bias must incorporate both technical safeguards and clinical workflow realities.
  • Data and infrastructure research needs to address scalability and integration barriers specific to healthcare environments.
  • New visualization methods are required to support explainability and communication between AI models and clinical users.
  • Human-AI interaction designs should be co-developed with domain experts to ensure clinical relevance.
  • Validation protocols for models must be adapted to the regulatory and safety demands of healthcare applications.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The clusters could serve as an organizing lens for grant programs that require cross-disciplinary teams.
  • Repeating the same synthesis process at future VAHC editions would reveal whether priorities shift over time.
  • Solutions developed for one cluster may transfer to others if common visualization techniques are identified.
  • Empirical deployments in actual hospital settings could test whether addressing these clusters measurably improves clinician adoption.

Load-bearing premise

Fifteen papers and one workshop edition with forty participants capture the central concerns of the broader visualization, AI, and healthcare communities.

What would settle it

A broader survey or follow-up workshop whose participants rank a materially different set of issues as highest priority would indicate that the five clusters do not represent community consensus.

read the original abstract

The intersection of AI, healthcare, and visualization is evolving rapidly, posing challenges that cut across disciplinary boundaries and resist easy resolution. The Visual Analytics in Healthcare workshop (VAHC), co-located every other year at the IEEE VIS conference and the AMIA (American Medical Informatics Association) annual conference, has served as a forum to connect the visualization and medical informatics community since 2010. In 2025, to celebrate the 16th edition, we used the workshop as an opportunity to consolidate the community's collective experience (and expertise) and identify Grand Challenges where the field should prioritize going forward. We combined thematic coding of the 15 accepted VAHC workshop papers with structured group discussions among more than 40 participants, organized around three major themes: "Technical innovation vs. clinical reality", "Human-centered and scalable VAHC", and "From foundations to actionable insights", followed by post-workshop reflexive analysis. Across all three groups, AI emerged as the most consistently recurring concern. In this paper, we report our AI-centered insights from the VAHC 2025 group activity, contextualize them against the broader literature along five Grand Challenges themes, and distill them into five challenge clusters, each concluded with recommendations for future research directions that cross disciplinary boundaries: (1) trust and bias, (2) data and infrastructure, (3) explainability and communication, (4) human-AI interaction, and (5) model reliability and validation. We share these challenges and their associated research directions as a starting point for discussion and collaboration across the healthcare, AI, and visualization communities. All supplemental materials are available at https://osf.io/p79uj.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 2 minor

Summary. The manuscript synthesizes AI-centered insights from the VAHC 2025 workshop by performing thematic coding on the 15 accepted papers and conducting structured group discussions among more than 40 participants organized around three themes ('Technical innovation vs. clinical reality', 'Human-centered and scalable VAHC', and 'From foundations to actionable insights'), followed by post-workshop reflexive analysis. It reports that AI was the most consistently recurring concern across groups, contextualizes the findings against broader literature along five Grand Challenges themes, and distills them into five challenge clusters—(1) trust and bias, (2) data and infrastructure, (3) explainability and communication, (4) human-AI interaction, and (5) model reliability and validation—each with recommendations for future cross-disciplinary research directions. The report positions the clusters as a starting point for discussion in the healthcare, AI, and visualization communities, with supplemental materials at https://osf.io/p79uj.

Significance. If the synthesis holds, the paper provides a timely, community-grounded snapshot of priorities at the AI-visualization-healthcare intersection. The combination of paper coding, multi-group discussions, and reflexive analysis offers a structured mechanism for capturing collective expertise from a single workshop edition, and the explicit acknowledgment of scope limitations (15 papers, one event, >40 participants) is a strength that prevents overgeneralization. The resulting five clusters with actionable recommendations can serve as a useful reference for directing collaborative research across the three fields.

minor comments (2)
  1. The abstract states that the five clusters were 'distilled' from the group activity plus literature contextualization, but the manuscript would benefit from a brief explicit mapping (e.g., in a methods or results subsection) showing how the three discussion themes contributed to each cluster; this would make the distillation process more transparent without altering the qualitative nature of the work.
  2. The supplemental materials URL is provided, yet the manuscript does not list the specific contents (e.g., full coding scheme, participant notes, or theme frequency counts); adding a short description of what is available would improve accessibility for readers wishing to build on the synthesis.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive assessment of the manuscript, the accurate summary of our methods and findings, and the recommendation for minor revision. No specific major comments were listed in the report.

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The paper is a descriptive synthesis of thematic coding from 15 accepted VAHC 2025 papers plus structured discussions among >40 participants, followed by post-workshop reflexive analysis and literature contextualization. The central claim—that AI emerged as the most recurring concern and was distilled into five challenge clusters—is a direct reporting of workshop input rather than any derivation, equation, fitted parameter, or prediction. No self-citation chains, uniqueness theorems, or ansatzes are invoked as load-bearing premises; the scope limitation to one workshop edition is stated explicitly. The output is therefore self-contained against external benchmarks with no reduction of claims to the paper's own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

No mathematical models, empirical measurements, or technical derivations are present. The paper rests on the domain assumption that workshop output can be treated as representative of field-wide priorities.

axioms (1)
  • domain assumption Thematic coding of 15 papers plus discussions with >40 participants yields a representative view of AI-centered challenges in visual analytics for healthcare.
    Invoked in the description of the synthesis process and the decision to distill into five clusters.

pith-pipeline@v0.9.1-grok · 5868 in / 1208 out tokens · 17972 ms · 2026-07-02T06:28:51.880051+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

53 extracted references · 26 canonical work pages

  1. [1]

    https: //visualanalyticshealthcare.github.io/homepage/2025/ call-for-papers.html

    Visual analytics in healthcare 2025 call for papers. https: //visualanalyticshealthcare.github.io/homepage/2025/ call-for-papers.html. Accessed: 2026-04-15. 1

    2025

  2. [2]

    E. S. Andersen, J. B. Birk-Korch, R. S. Hansen, L. H. Fly, et al. Monitoring performance of clinical artificial intelligence in health care: a scoping review.JBI Evidence Synthesis, 22(12):2423–2446, 2024. doi: 10.11124/ JBIES-24-00042 4

    2024

  3. [3]

    Arleo, R

    A. Arleo, R. Borgo, J. Kohlhammer, R. A. Ruddle, H. Scharlach, and X. Yuan. Reflections on the use of dashboards in the COVID-19 pandemic. IEEE Computer Graphics and Applications, 45(2):135–142, 2025. doi: 10 .1109/MCG.2025.3538257 2

    work page arXiv 2025

  4. [4]

    Arleo, A

    A. Arleo, A. T. Chen, D. Gotz, S. Kandaswamy, and J. Bernard. Re- flections on interactive visualization of electronic health records: past, present, future.Journal of the American Medical Informatics Association, 31(11):2423–2428, 10 2024. doi: 10.1093/jamia/ocae249 1

    work page doi:10.1093/jamia/ocae249 2024

  5. [6]

    Assor, H

    A. Assor, H. Jeon, S. Shin, and J.-D. Fekete. Changing the paradigm from dynamic queries to LLM-generated SQL queries with human interven- tion. InVIS Workshop on Visual Analytics in Healthcare (VAHC). IEEE,

  6. [7]

    [Online] https://visualanalyticshealthcare.github.io/ homepage/proceedings/2025/vis25n-sub6788-cam-i7.pdf. 2, 3

    2025

  7. [8]

    Aupetit and A

    M. Aupetit and A. Abuthawabeh. InViTAG: A web application for AI-assisted exploration and grouping of health images and data. InVIS Workshop on Visual Analytics in Healthcare (VAHC). IEEE,

  8. [9]

    [Online] https://visualanalyticshealthcare.github.io/ homepage/proceedings/2025/vis25n-sub6439-cam-i7.pdf. 3

    2025

  9. [11]

    Baumgartl, M

    T. Baumgartl, M. Ghoniem, T. von Landesberger, G. E. Marai, S. Miksch, S. Mohr et al. Empowering communities: Tailored pandemic data vi- sualization for varied tasks and users.IEEE Computer Graphics and Applications, 45(1):130–138, 2025. doi: 10.1109/MCG.2024.3509293 3

    work page doi:10.1109/mcg.2024.3509293 2025

  10. [12]

    Bernard, D

    J. Bernard, D. Borland, R. S. Laramee, B. Preim, R. G. Raidou, and A. Vi- lanova. Future challenges and unsolved problems in health visualization. InEuroVis 2025 - Panels and Tutorials. Eurographics, 2025. doi: 10. 2312/evt.20251119 1

    2025

  11. [13]

    Brückner, O

    S. Brückner, O. Sadare, S. Fesl, M. Scheibe, C. Lang, and S. Gilbert. Attitudes of healthcare professionals and researchers toward wearable and app derived patient generated health data.npj Digital Medicine, 8(1):186, Mar. 2025. doi: 10.1038/s41746-025-01568-4 3

    work page doi:10.1038/s41746-025-01568-4 2025

  12. [14]

    Buhler, T

    K. Buhler, T. Hollt, T. Schultz, and P.-P. Vazquez. AI-in-the-Loop: The future of biomedical visual analytics applications in the era of AI.IEEE Computer Graphics and Applications, 2025. doi: 10.1109/MCG.2024. 3517293 3

    work page doi:10.1109/mcg.2024 2025

  13. [15]

    J. J. Caban and D. Gotz. Visual analytics in healthcare - opportunities and research challenges.J. Am. Medical Informatics Assoc., 22(2):260–262,

  14. [16]

    doi: 10.1093/JAMIA/OCV006 1

    work page doi:10.1093/jamia/ocv006

  15. [17]

    R. J. Chen, J. J. Wang, D. F. K. Williamson, T. Y . Chen, J. Lipkova, M. Y . Lu et al. Algorithmic fairness in artificial intelligence for medicine and healthcare.Nature Biomedical Engineering, 7(6):719–742, 2023. doi: 10. 1038/s41551-023-01056-8 2

    2023

  16. [18]

    J. L. Cross, M. A. Choma, and J. A. Onofrey. Bias in medical AI: Implica- tions for clinical decision-making.PLOS Digital Health, 3(11):e0000651,

  17. [19]

    doi: 10.1371/journal.pdig.0000651 2

    work page doi:10.1371/journal.pdig.0000651

  18. [20]

    Ershova and G

    M. Ershova and G. Blasilli. Bridging service design, visualiza- tions, and visual analytics in healthcare digital twins: Chal- lenges, gaps, and research opportunities. InVIS Workshop on Visual Analytics in Healthcare (VAHC). IEEE, 2025. [Online] https://visualanalyticshealthcare.github.io/homepage/ proceedings/2025/vis25n-sub6333-cam-i7.pdf. 2, 3

    2025

  19. [21]

    S. G. Finlayson, A. Subbaswamy, K. Singh, J. Bowers, A. Kupke, J. Zit- train et al. The clinician and dataset shift in artificial intelligence.New England Journal of Medicine, 385(3):283–286, 2021. doi: 10.1056/ NEJMc2104626 4

    2021

  20. [22]

    B. A. Ghassemi M, Oakden-Rayner L. The false hope of current ap- proaches to explainable artificial intelligence in health care.The Lancet Digital Health, 3(11):e745–e750, 2021. doi: 10.1016/S2589-7500(21) 00208-9 3

    work page doi:10.1016/s2589-7500(21 2021

  21. [23]

    S. Ha, S. Monadjemi, and A. Ottley. Guided by ai: Navigating trust, bias, and data exploration in ai-guided visual analytics.Computer Graphics Forum, 43(3):e15108, 2024. doi: 10.1111/cgf.15108 2

    work page doi:10.1111/cgf.15108 2024

  22. [24]

    Hosseini, J

    A. Hosseini, J. Wood, A. Slingsby, R. Jianu, and M. Elshehaly. What do LLMs prioritise when adapting visualizations to user personas? In Workshop on Visual Analytics in Healthcare (VAHC), pp. 15–21. IEEE,

  23. [26]

    Kleinau, J

    A. Kleinau, J. B. Au, J. Ulmer, E. Charbonnier, M. Y . Jabarulla, S. E. Long et al. A dashboard for visualizing health data of children with aortic valve disease. InVIS Workshop on Visual Analytics in Healthcare (VAHC). IEEE,

  24. [27]

    [Online] https://visualanalyticshealthcare.github.io/ homepage/proceedings/2025/vis25n-sub8649-cam-i7.pdf. 3

    2025

  25. [28]

    V . Lai, C. Chen, A. Smith-Renner, Q. V . Liao, and C. Tan. Towards a science of human-AI decision making: An overview of design space in empirical human-subject studies. InACM Conference on Fairness, Accountability, and Transparency (FAccT), pp. 1369–1385. Association for Computing Machinery, 2023. doi: 10.1145/3593013.3594087 4

    work page doi:10.1145/3593013.3594087 2023

  26. [30]

    J. K. Lennerz, U. Green, D. F. K. Williamson, and F. Mahmood. A unifying force for the realization of medical AI.npj Digital Medicine, 5(1):172,

  27. [31]

    doi: 10.1038/s41746-022-00721-7 4

    work page doi:10.1038/s41746-022-00721-7

  28. [32]

    Meske, J

    C. Meske, J. Brenne, E. Uenal, S. Oelcer, and A. Doganguen. From explainable to explanatory artificial intelligence: Toward a new paradigm for human-centered explanations through generative ai.arXiv preprint arXiv:2508.06352, 2025. 3

    work page arXiv 2025

  29. [33]

    Morgenshtern, A

    G. Morgenshtern, A. Verma, S. Tonekaboni, R. Greer, J. Bernard, M. Mazwi et al. RiskFix: Supporting Expert Validation of Predictive Timeseries Models in High-Intensity Settings. InEuroVis 2023 (Short). Eurographics, 2023. doi: 10.2312/evs.20231036 4

    work page doi:10.2312/evs.20231036 2023

  30. [34]

    Musleh, R

    M. Musleh, R. G. Raidou, and D. Ceneda. Trustme: A context-aware explainability model to promote user trust in guidance.IEEE Transactions on Visualization and Computer Graphics, 2025. 2

    2025

  31. [36]

    Owen and J

    R. Owen and J. C. Roberts. Embedding empathy into visual analytics: A framework for person-centred dementia care. InWorkshop on Visual Analytics in Healthcare (VAHC), pp. 43–49. IEEE, 2025. doi: 10.1109/ V AHC69430.2025.00012 3

    work page arXiv 2025

  32. [37]

    P. V . S. Pakianathan, R. Islambouli, H. McGowan, D. Branco, T. Guer- reiro, and J. D. Smeddinck. Exploring human-AI interaction with patient-generated health data sensemaking for cardiac risk reduction. InVIS Workshop on Visual Analytics in Healthcare (VAHC). IEEE,

  33. [38]

    [Online] https://visualanalyticshealthcare.github.io/ homepage/proceedings/2025/vis25n-sub4125-cam-i7.pdf. 3

    2025

  34. [39]

    Queralt-Rosinach, R

    N. Queralt-Rosinach, R. Kaliyaperumal, C. H. Bernabé, Q. Long, S. A. Joosten, H. J. van der Wijk et al. Applying the FAIR principles to data in a hospital: challenges and opportunities in a pandemic.Biomedical Semantics, 13(1):12, 2022. doi: 10.1186/s13326-022-00263-7 3

    work page doi:10.1186/s13326-022-00263-7 2022

  35. [40]

    Rauscher, M

    J. Rauscher, M. Schmidt, U. Schlegel, M. Bieleke, J. Schüler, and D. A. Keim. MotiV Ator: Analyzing physical activity study data in lab and life through visual analytics. InWorkshop on Visual Analytics in Healthcare (VAHC), pp. 8–14. IEEE, 2025. doi: 10.1109/V AHC69430.2025.00006 3

    work page doi:10.1109/v 2025

  36. [41]

    S. Reardon. Rise of Robot Radiologists.Nature, 576(7787):S54–S58, Dec. 2019. doi: 10.1038/d41586-019-03847-z 1

    work page doi:10.1038/d41586-019-03847-z 2019

  37. [42]

    Ristovski, T

    G. Ristovski, T. Preusser, H. K. Hahn, and L. Linsen. Uncertainty in medical visualization: Towards a taxonomy.Computers & Graphics, 39:60–73, 2014. doi: 10.1016/j.cag.2013.10.015 3

    work page doi:10.1016/j.cag.2013.10.015 2014

  38. [43]

    C. Rudin. Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead.Nature Machine Intelligence, 1(5):206–215, 2019. doi: 10.1038/s42256-019-0048-x 3

    work page doi:10.1038/s42256-019-0048-x 2019

  39. [44]

    Sacha, H

    D. Sacha, H. Senaratne, B. C. Kwon, G. Ellis, and D. A. Keim. The role of uncertainty, awareness, and trust in visual analytics.IEEE Transactions on Visualization and Computer Graphics, 22(1):240–249, 2016. doi: 10. 1109/TVCG.2015.2467591 2

    work page arXiv 2016

  40. [45]

    Shang, V

    Z. Shang, V . Chauhan, K. Devi, and S. Patil. Artificial Intelligence, the Digital Surgeon: Unravelling Its Emerging Footprint in Healthcare – The Narrative Review.Journal of Multidisciplinary Healthcare, V olume 17:4011–4022, Aug. 2024. doi: 10.2147/JMDH.S482757 1

    work page doi:10.2147/jmdh.s482757 2024

  41. [46]

    Shell, A

    J. Shell, A. K. Davis, J. Bai, and J. Aurisano. Explaining model pre- dictions when users aren’t sure: Xai for scrna-seq cell-type classifica- tion. InVIS Workshop on Visual Analytics in Healthcare (VAHC). IEEE,

  42. [47]

    [Online] https://visualanalyticshealthcare.github.io/ homepage/proceedings/2025/vis25n-sub3665-cam-i7.pdf. 3, 4

    2025

  43. [48]

    Singh, R

    S. Singh, R. Kumar, S. Payra, and S. K. Singh. Artificial Intelligence and Machine Learning in Pharmacological Research: Bridging the Gap Between Data and Drug Discovery.Cureus, Aug. 2023. doi: 10.7759/ cureus.44359 1

    2023

  44. [49]

    Stoiber, D

    C. Stoiber, D. Ceneda, M. Wagner, V . Schetinger, T. Gschwandtner, M. Streit et al. Perspectives of visualization onboarding and guidance in va.Visual Informatics, 6(1):68–83, 2022. doi: 10.1016/j.visinf.2022.02. 005 4

    work page doi:10.1016/j.visinf.2022.02 2022

  45. [50]

    A. Suh, G. Appleby, E. W. Anderson, L. Finelli, R. Chang, and D. Cash- man. Are metrics enough? guidelines for communicating and visualizing predictive models to subject matter experts.IEEE Transactions on Vi- sualization and Computer Graphics, 30(7):4137–4153, 2024. doi: 10. 1109/TVCG.2023.3259341 4

    work page arXiv 2024

  46. [51]

    S. Vink, K. Byungmoo, C. Brumar, M. Yang, K. Potter, and R. Chang. Position Paper: The Opportunity for Visual Analytics in the Age of Gener- ative AI. In A. Villanova and D. Archambault, eds.,EuroVA 2026: EuroVis Workshop on Visual Analytics. The Eurographics Association, 2026. doi: 10.2312/eurova.20261014 4

    work page doi:10.2312/eurova.20261014 2026

  47. [52]

    von Wyl and J

    V . von Wyl and J. Bernard. Grand challenge: The ur- gent need for methods and tools to mediate between the re- search cultures of health sciences and visual analytics. In VIS Workshop on Visual Analytics in Healthcare (VAHC). IEEE,

  48. [53]

    [Online] https://visualanalyticshealthcare.github.io/ homepage/proceedings/2025/vis25n-sub9654-cam-i7.pdf. 2, 3

    2025

  49. [54]

    A. Z. Wang, D. Borland, and D. Gotz. Visual analytics for causal reasoning from real-world health data. InVIS Workshop on Visual Analytics in Healthcare (VAHC). IEEE, 2025. [Online] https://visualanalyticshealthcare.github.io/homepage/ proceedings/2025/vis25n-sub1773-cam-i7.pdf. 2, 3

    2025

  50. [55]

    Wang and R

    Q. Wang and R. Laramee. Ehr star: The state-of-the-art in interactive ehr visualization.Computer Graphics Forum, 41(1):69–105, 2022. doi: 10. 1111/cgf.14424 3

    2022

  51. [56]

    M. D. Wilkinson, M. Dumontier, I. J. Aalbersberg, G. Appleton, M. Axton, A. Baak et al. The FAIR guiding principles for scientific data management and stewardship.Scientific Data, 3:160018, 2016. doi: 10.1038/sdata. 2016.18 3

    work page doi:10.1038/sdata 2016

  52. [57]

    K. Xu, A. Ottley, C. Walchshofer, M. Streit, R. Chang, and J. Wenskovitch. Survey on the analysis of user interactions and visualization provenance. Computer Graphics Forum, 39(3):757–783, 2020. doi: 10.1111/cgf.14035 3

    work page doi:10.1111/cgf.14035 2020

  53. [58]

    S. Xu, T. Rogers, E. Fairweather, A. Glenn, J. Curran, and V . Curcin. Appli- cation of Data Provenance in Healthcare Analytics Software: Information Visualisation of User Activities.AMIA Joint Summits on Translational Science proceedings, 2017:263–272, 2018. 3

    2017