The NHS ADOPT study has now begun to automatically identify patients with vertebral fracture, by using Nanox.AI an Artificial Intelligence (AI) programme to review thousands of hospital computed tomography (CT) scans. The patients are now being fast-tracked for bone health assessment by an NHS osteoporosis team.
When a person breaks a bone after the age of 50, it can be the first sign of osteoporosis. Until a fracture occurs, osteoporosis is a silent but treatable condition which causes the bones to become more porous and fragile. The challenge is to find people at risk. Early diagnosis is key, since simple, effective osteoporosis treatments can prevent the misery of broken bones. Untreated, osteoporosis can cause a curved back and progressive height loss. While breaking an arm or a leg is obvious, many people have breaks of their spine that are ignored as 'ordinary' back pain and therefore go undiagnosed. The ADOPT study researchers are looking at a novel way to automatically find these spine fractures in people attending hospital for ordinary diagnostic CT scans by utilising Nanox.AI solution.
ADOPT is a collaboration between the University of Oxford, Addenbrooke's Hospital, Cambridge, medical imaging technology company Nanox.AI, and the Royal Osteoporosis Society. The study team were funded and supported by the National Institute for Health and Care Research (NIHR) and NHS England to test an innovative AI programme that automatically 'reads through' CT scan pictures to find spine fractures. Each year in the UK, more than three million people have CT scans that include the spine. Since these scans are done for other reasons (e.g. lung, bowel or for monitoring various diseases) the doctors reading or 'reporting' the scans sometimes have their attention focused on those other areas. This means that in many hospitals spine fractures aren't identified or reported, even though more than 1 in 20 CT scans do feature an osteoporotic spine fracture.
Nanox.AI HealthVCF Artificial Intelligence programme reviews all local CT scans to find spine fractures and brings them directly to the specialist team's attention. It has just identified the first cases of suspected vertebral fracture. Daniel Chappell, Clinical Projects manager in Addenbrooke's Hospital Cambridge explained: 'Within the first month of my switching on the Nanox.AI-fracture detection system, several thousand CT scans were processed. From 497 alerts made in July 2023, 244 (49%) correctly identified patients with true vertebral fractures. Examining the records of 244 patients, 170 of them required osteoporosis assessment/treatment. While 147 were already scheduled for bone health management using our current spine fracture detection pathways, a total of 23 people found by Nanox.AI's HealthVCF are now scheduled for a bone health assessment and hopefully fracture-preventing treatment.'
Professor Kassim Javaid, leading the research at the Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences (NDORMS), University of Oxford said: 'Finding and assessing the first patient from the ADOPT study for osteoporosis to improve their bone health and reduce their risk of fracture represents a milestone for the study. We hope the positive benefits from using Nanox.AI will significantly change the lives of thousands of patients in the project and potentially hundreds of thousands of lives in the NHS. It also represents the fruition of the combined efforts from clinicians, researchers, IT experts and the teams from Nanox and the Royal Osteoporosis Society, including patients, to implement this novel care pathway into the NHS. We now eagerly await the results of the study in 2024.'
If successful, an AI-enhanced vertebral fracture prevention pathway could improve identification, assessment, and treatment recommendation for osteoporosis patients and offer potential cost savings for the NHS.
The study is funded by the National Health Service England (NHSE) and the National Institute for Health and Care Research (NIHR), the research partner of the NHS, public health and social care. This study was supported by the NIHR Cambridge Biomedical Research Centre.
