Integrating AI into clinical practice: Human-centered design requirements for next-generation sequencing workflows

Area of Science:

• Genomics
• Bioinformatics
• Clinical Informatics

Background:

• Next-generation sequencing (NGS) is crucial for clinical genomics, but its integration is challenged by fragmented workflows and usability issues.
• Artificial intelligence (AI) is increasingly driving NGS analysis, from data processing to clinical decision support.

Purpose of the Study:

• To present a design-oriented framework (DUXU) for embedding AI-powered NGS workflows into clinical decision support systems (CDSS).
• To address the socio-technical demands of clinical genomics and propose actionable design requirements for AI-based systems.

Main Methods:

• Development of a conceptual and methodological framework (DUXU) focused on Design, User eXperience, and Usability.
• Grounded in real-world clinical environments and aligned with standards like FHIR and GA4GH.
• Focus on AI's role in multimodal data interpretation, patient-specific visualization, and explainable decision-making.

Main Results:

• The DUXU framework offers a flexible approach adaptable to specific clinical use cases, including genetic screening, tumor testing, and potential pathogen detection.
• Highlights the central role of AI in creating trustworthy, interpretable, and operationally embedded AI-based NGS systems.
• Proposes actionable design requirements for interoperable, role-specific interfaces.

Conclusions:

• The DUXU framework facilitates the integration of AI-driven NGS into clinical practice by addressing usability and workflow challenges.
• Advances the development of explainable AI for clinical genomics, enhancing diagnostic and treatment strategies.
• Future work will extend the framework to diverse applications like pathogen detection and antimicrobial stewardship.