A Domain-Oriented Approach to Designing an AI-Supported PCD Diagnostic Tool

Maria Fedosenya

Supervisor(s): Ing. Martin Dubovský

Slovak Technical University

Abstract: Primary ciliary dyskinesia (PCD) is a rare genetic disorder that requires a multimodal diagnostic approach and a close collaboration between domain experts. While artificial intelligence (AI) has the potential to enhance medical decision-making, the successful integration of AI into clinical diagnostics depends on careful alignment with domain-specific workflows and expert needs. This paper presents a domain-oriented, co-design methodology for developing diagnostic tools in specialized medical domains. Our research is grounded in domain-oriented user research, including semi-structured interviews, clinical workflow analysis, and participatory design sessions with PCD experts. During the design process we adapted the Double Diamond model to ensure the continuous domain knowledge transfer and enable the involvement of the domain experts throughout the concept development and prototyping phase. The resulting prototypes address identified gaps in the clinical workflow, the specifics and limitations of the domain. The proposed interface is designed to support the enhancement of collaboration among experts and integration of AI into existing clinical workflows. The work contributes a structured design methodology and provides practical insights into designing domain-specific diagnostic systems, demonstrated through a case study in PCD diagnostics.
Keywords: Design, Human-Computer Interaction
Full text:
Year: 2026