English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Details of the designs and mechanisms in support of human-AI collaborationmust be considered in the real-world fielding of AI technologies. A criticalaspect of interaction design for AI-assisted human decision making are policiesabout the display and sequencing of AI inferences within larger decision-makingworkflows. We have a poor understanding of the influences of making AIinferences available before versus after human review of a diagnostic task athand. We explore the effects of providing AI assistance at the start of adiagnostic session in radiology versus after the radiologist has made aprovisional decision. We conducted a user study where 19 veterinaryradiologists identified radiographic findings present in patients' X-rayimages, with the aid of an AI tool. We employed two workflow configurations toanalyze (i) anchoring effects, (ii) human-AI team diagnostic performance andagreement, (iii) time spent and confidence in decision making, and (iv)perceived usefulness of the AI. We found that participants who are asked toregister provisional responses in advance of reviewing AI inferences are lesslikely to agree with the AI regardless of whether the advice is accurate and,in instances of disagreement with the AI, are less likely to seek the secondopinion of a colleague. These participants also reported the AI advice to beless useful. Surprisingly, requiring provisional decisions on cases in advanceof the display of AI inferences did not lengthen the time participants spent onthe task. The study provides generalizable and actionable insights for thedeployment of clinical AI tools in human-in-the-loop systems and introduces amethodology for studying alternative designs for human-AI collaboration. Wemake our experimental platform available as open source to facilitate futureresearch on the influence of alternate designs on human-AI workflows.

Who Goes First? Influences of Human-AI Workflow on Decision Making in Clinical Imaging