A Manchester University scientist has won funding from a medical-research charity to develop a more effective way of routinely detecting osteoporosis in the spine.
Tim Cootes, professor of computer vision in the School of Cancer and Enabling Sciences, has been awarded £136,600 over two years from Arthritis Research UK to carry out the work.
Working with Prof Judith Adams at Manchester Royal Infirmary, Prof Cootes and his team aim to produce a new computer program to detect fractured vertebrae in radiographs, which are commonly used to diagnose osteoporosis. Currently, up to 50 per cent of fractured vertebrae are missed, as identifying them requires considerable expertise.
‘The tool will automatically measure the shape of each vertebra and indicate if it is damaged or unusual, so helping clinicians to detect vertebral fractures more reliably and quickly,’ said Cootes. ‘This will help to diagnose osteoporosis at an earlier stage, leading to earlier treatment and reducing the risk of future fractures.’
Osteoporosis is a condition affecting the bones, causing them to weaken and fracture. Around three million people in the UK have the disease — mainly post-menopausal women — leading to more than 230,000 fractures. Lifestyle changes and drug treatments can reduce the risk of fractures, so early diagnosis is important.
An early sign of the disease is fracture of the vertebrae, but because fractures often cause no discomfort they can go unnoticed by the patient.
‘We’ve developed a computer system that can distinguish between normal and fractured vertebrae, and in this project we aim to improve the system further and to train it to work on the different types of X-ray images that are commonly used,’ said Cootes.
‘We’ll examine routine clinical images to test if the tool is reliable enough to help clinicians in normal examinations.’