July 28, 2025 — Copenhagen, Denmark — Faced with rising demand and stretched radiology resources, Lewisham and Greenwich NHS Trust embarked on a mission to integrate AI-powered fracture detection into its emergency services. The Trust adopted a rigorous evaluation framework aligned with BS 30440, the British Standard for validating AI in healthcare, to compare multiple solutions before procurement.
As part of this initiative, the Trust conducted retrospective and shadow mode testing of two fracture detection AIs, alongside a comprehensive clinical safety report and ethical impact assessment using the Z-Inspection® Framework. The process included structured performance metrics, stakeholder engagement, and strong governance mechanisms to ensure safe and effective integration into the clinical environment.
By using the deepcOS® AI Infrastructure Platform from deepc, Lewisham was able to independently evaluate multiple AI solutions in a structured, BS 30440-aligned process, selecting RBfracture™ as the best fit for their fracture detection workflow.
Following the BS 30440-aligned evaluation, RBfracture™ demonstrated superior performance, achieving 97% accuracy, 93% sensitivity, and 98.8% specificity compared with the other AI solution tested. The Trust concluded that the standard-driven approach improved the quality, transparency, and safety of its fracture detection processes.
What BS 30440, Validation framework for the use of AI within healthcare, covers
BS 30440:2023 provides a structured, auditable validation framework for AI in healthcare, defining how systems should be evaluated for clinical benefit, performance, safe integration, ethical impact, and social equity. It establishes a consistent and evidence-based methodology that healthcare providers can use to assess AI products before deployment, reducing uncertainty and improving trust in AI-assisted care.
Why does using a framework like BS 30440 matter?
BS 30440 gives hospitals and AI developers a shared language for assessing and demonstrating clinical value. For healthcare providers, it means more confidence in selecting AI tools that meet defined safety and efficacy standards. For vendors, it ensures transparency and accountability in how performance and ethical considerations are reported.
By adopting this framework, Lewisham and Greenwich NHS Trust has set a benchmark for how AI can be responsibly and effectively introduced into clinical workflows—supporting clinicians, improving accuracy, and ultimately enhancing patient care.
“Applying the BS 30440 framework in our comparative AI study has been transformative: not only in selecting the most clinically viable fracture detection tool for our service, but also in aligning our adoption process with safety, ethical, and regulatory best practices. The structured evaluation has strengthened confidence and provides excellent guidance and support to anyone embarking on the deployment of NHS AI applications, where the AI technologies are still evolving.”
Dr Sarojini David
Clinical Director of Radiology, AI Lead
Lewisham and Greenwich NHS Trust
Download the comparative assessment
BS 30440 AI Fracture Detection Framework
This original research was conducted by Lewisham and Greenwich NHS Trust.
Entering your information does not subscribe you to marketing emails from Radiobotics. You can always read our privacy policy.
