If Clinicians Designed AI: Creating Seamless, Human-Centered Imaging Solutions

As AI portfolios in imaging continue to expand, many organizations find that deployment is moving faster than user education, workflow alignment, and thoughtful interface design. Attendees will explore why technically strong AI models often fail to deliver real efficiency gains when results are difficult to access, inconsistently presented, or poorly aligned with clinical workflows. The session establishes the core challenge: without intentional design that accounts for cognitive load, alert fatigue, and clinician expectations, AI can unintentionally worsen the day-to-day user experience rather than improve it. 

Participants will learn practical strategies for designing and evaluating AI result presentation so that outputs fit naturally into imaging workflows. The session focuses on aligning AI interfaces with radiologists’ mental models, prioritizing critical findings, minimizing unnecessary clicks, and avoiding information overload. Attendees will gain insight into how thoughtful UI/UX decisions, use of standards, and exam-status awareness can improve trust, reduce friction, and support consistent clinical adoption. Real-world examples will illustrate how poor design choices lead to missed results, delayed interpretation, and alert fatigue—and how these issues can be mitigated through prospective user involvement. 

Building on these principles, participants will examine a real-world case study demonstrating how AI can be embedded into secure, auditable, and trusted imaging workflows. Through examples drawn from hybrid human–AI implementations, attendees will learn how containerized models, rule-based orchestration, metadata tagging, and human review can work together to support regulatory compliance and clinical trust. The session emphasizes transferable lessons across imaging domains, giving attendees practical takeaways they can apply when working with vendors, developers, and internal teams to improve AI integration, usability, and adoption.  

Objectives 

• Identify common UI/UX design pitfalls that increase cognitive and physical burden when AI results are introduced into imaging workflows. 
• Explain how standards, workflow alignment, and exam-status awareness improve usability, trust, and clinical adoption of AI solutions. 
• Implement practical strategies for integrating AI into secure, auditable, and user-centered imaging workflows that balance automation with human oversight.