top of page

AI to Revolutionize X-Ray Fracture Detection: NICE Recommends New AI Tools for Urgent Care

Oct 22

3 min read

0

4

0

Artificial intelligence (AI) is set to play a significant role in reducing the number of missed broken bones on X-rays, according to recommendations by the National Institute for Health and Care Excellence (NICE). With healthcare professionals under immense pressure and facing staff shortages, AI-powered tools are being considered as a solution to support clinicians in urgent care settings.


Implementing AI technologies aims to help doctors identify fractures more effectively. Four AI tools—TechCare Alert, BoneView, RBfracture, and Rayvolve—are expected to be used alongside healthcare professionals in urgent care settings in England. The AI-powered tools will collaborate, as a trained radiologist or radiographer will still review each X-ray. NICE has indicated that these AI systems are safe, and their integration into medical practice could enhance the speed and accuracy of fracture detection, ultimately relieving the pressure on overburdened healthcare workers.

Research indicates that AI could significantly reduce the number of broken bones missed during initial X-ray analysis. Missed fractures are the most common diagnostic error in emergency departments, with 3% to 10% of cases involving undetected breaks. This new guidance from NICE suggests that AI could help mitigate this issue. The draft guidance, released on October 22, 2024, recommends implementing these AI technologies in urgent care while further evidence is collected to demonstrate their full benefits in real-world applications.


The healthcare industry faces high vacancy rates in radiology and radiography, with 12.5% of radiologist positions and 15% of radiographer positions currently unfilled in England. These staffing shortages increase the workload for existing professionals, which can lead to mistakes or missed diagnoses. Mark Chapman, the director of health technology at NICE, emphasized that AI could support these professionals by spotting fractures that might be overlooked due to the demanding nature of their work. He further explained that using AI could speed up diagnoses and reduce the number of follow-up appointments caused by missed fractures during initial assessments.

NICE is confident that AI-assisted fracture detection will not increase unnecessary referrals to fracture clinics, given that a healthcare professional will always be involved in reviewing the X-rays. AI may enhance care by reducing variations in treatment across different regions and helping ensure that more fractures are identified early, potentially preventing further harm or complications due to delayed diagnosis.


While the AI tools already demonstrate potential benefits, the draft guidance is still subject to consultation. NICE is gathering feedback from stakeholders until November 5, 2024, after which the committee will review the input and potentially update its recommendations. Final guidance is expected to be published by January 14, 2025.

This move toward AI-assisted healthcare follows broader trends in the sector, where AI is increasingly being used to detect early signs of conditions such as breast cancer and heart attacks and even predict future pandemics. Additionally, the UK government has taken steps to support the safe deployment of AI innovations in healthcare. In October 2024, Science and Technology Secretary Peter Kyle announced the launch of a new regulatory office dedicated to ensuring that AI technologies are used safely and effectively within the healthcare system.


As AI continues to evolve, its role in healthcare is expected to expand. It offers the potential to improve diagnostic accuracy, streamline processes, and alleviate some of the burdens placed on healthcare professionals. With AI poised to transform fracture detection and other areas of healthcare, the upcoming NICE guidance will be a crucial step in integrating these technologies into clinical practice.


Comments

Share Your ThoughtsBe the first to write a comment.
bottom of page