Answers are grounded in this paper and cite their sources.
Focus groups revealed improved usability and patient experience of AI-assisted capillaroscopy, suggesting enhanced systemic sclerosis diagnostics.
Background/Aims Despite the diagnostic value of nailfold capillaroscopy, access is limited due to a shortage of trained personnel and the technical demands of image acquisition and interpretation. Artificial intelligence (AI) presents an opportunity to expand access by supporting non-specialists in capturing and interpreting images through automated systems. We developed an AI-guided system that integrates automated image capture, real-time feedback, and reporting capabilities. This study aimed to gather insights from healthcare professionals and patients to assess the system’s acceptability, usability, and sustainability. Methods We conducted online focus groups with participants from a pilot study involving AI-guided nailfold capillaroscopy. The sample included six patients with systemic sclerosis and eight healthcare professionals across three NHS hospitals. Discussions were held virtually and audio-recorded, transcribed, and analysed using a framework approach. Themes explored included training needs, image acquisition, reporting clarity, integration into clinical workflows, and patient experience. Additionally, questionnaires were distributed to those participating in the pilot study (n = 13 staff; N = 87 patients) to evaluate usability, tolerability, and suggestions for improvement. Patient and healthcare professional involvement informed the study’s design and delivery. Ethical approval was obtained, and further focus groups are planned for Autumn 2025 to assess iterative refinements. Results Preliminary findings revealed strong support for the system’s potential to reduce time, travel, and associated costs. Patients described varied diagnostic pathways, often delayed by limited recognition of systemic sclerosis in primary care. Nailfold capillaroscopy was well tolerated, with 98.8% of patients agreeing or strongly agreeing they felt physically comfortable. Our nailfold capillaroscopy software was seen as a progressive step compared to traditional diagnostic methods. Both clinicians and patients welcomed the AI-supported system, provided clinical interpretation remained clinician-led. Among staff, 59% agreed or strongly agreed that the software was sufficiently intuitive to complete tasks. However, those new to the system reported long image acquisition times during initial use (38% agreed or strongly agreed they completed tasks quickly). Automated reports were positively received, though feedback suggested refinements to layout and language to aid interpretation by non-specialists - particularly around image quality indicators, and reference ranges tailored to specific populations. Practical considerations such as Wi-Fi reliability and clinic space also impacted implementation. Conclusion AI-assisted nailfold capillaroscopy offers a promising route to modernise and facilitate access to systemic sclerosis diagnostics, enhancing both efficiency and equity. User feedback identified actionable improvements in software design and reporting functionality. The findings also highlighted the need for broader service redesign to support adoption - addressing not only technical aspects but also patient experience and diagnostic pathways (e.g. longer initial appointments to accommodate capillaroscopy). Future user engagement will inform the next iteration of the system and guide implementation strategies.
No takes yet. Share an insight, caveat, or question.
Wilkinson et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: