AI to assist doctors in detecting broken bones in X-rays

Artificial intelligence (AI) has the potential to reduce the number of missed broken bones in X-rays, according to the National Institute for Health and Care Excellence (NICE). Research indicates that this technology is safe and could accelerate diagnosis, easing the burden on clinicians and minimizing the need for follow-up appointments.

NICE plans to recommend four AI tools for urgent care in England while further evidence on their benefits is gathered. Importantly, AI will complement the work of healthcare professionals, as each X-ray will still be reviewed by a qualified expert.

NICE notes that broken bones are overlooked in 3-10% of cases, making it the most common diagnostic error in emergency departments. The NHS faces a shortage of trained specialists, with vacancy rates at 12.5% for radiologists and 15% for radiographers, contributing to overwhelming workloads.

Mark Chapman, director of health technology at NICE, stated that AI could ease the strain on professionals. “These AI technologies are safe to use and could identify fractures that may be overlooked, given the pressure these professionals face,” he explained. He also suggested that AI could expedite diagnoses and reduce the number of follow-up visits for missed fractures.

NICE believes it is “unlikely” that the technology will increase incorrect diagnoses or unnecessary referrals, as a radiologist will always review the images. This approach is expected to be more effective than having a clinician analyze the images alone.

The potential applications of AI in healthcare are vast, with current uses including early detection of breast cancer, identifying individuals at high risk for heart attacks, and predicting future pandemics.

A consultation on the draft NICE guidance regarding the use of AI in this context will close on November 5.