An algorithm to monitor the joints of patients witharthritis, which could change the way that the severity of the condition is assessed, has been developedby a team of engineers, physicians and radiologistsled by the University of Cambridge.
The technique, which detects tiny changes in arthriticjoints, could enable greater understanding of how osteoarthritis develops and allow the effectiveness ofnew treatments to be assessed more accurately, without the need for invasive tissue sampling. The resultsare published in the journal Scientific Reports.
Osteoarthritis is the most common form of arthritis inthe UK. It develops when the articular cartilage thatcoats the ends of bones, and allows them to glidesmoothly over each other at joints, is worn down, resulting in painful, immobile joints. Currently there isno recognised cure and the only definitive treatmentis surgery for artificial joint replacement.
Osteoarthritis is normally identified on an x-ray by anarrowing of the space between the bones of the jointdue to a loss of cartilage. However, x-rays do not haveenough sensitivity to detect subtle changes in the jointover time. "In addition to their lack of sensitivity, two-dimensional x-rays rely on humans to interpret them," saidlead author Dr Tom Turmezei from Cambridge's Department of Engineering. "Our ability to detect structural changes to identify disease early, monitorprogression and predict treatment response is frustratingly limited by this."
The technique developed by Turmezei and his colleagues uses images from a standard computerisedtomography (CT) scan, which isn't normally used tomonitor joints, but produces detailed images in threedimensions.
The semi-automated technique, called joint spacemapping (JSM), analyses the CT images to identifychanges in the space between the bones of the joint inquestion, a recognised surrogate marker for osteoarthritis. After developing the algorithm with testson human hip joints from bodies that had been donated for medical research, they found that it exceeded the current 'gold standard' of joint imagingwith x-rays in terms of sensitivity, showing that it wasat least twice as good at detecting small structuralchanges. Colour-coded images produced using theJSM algorithm illustrate the parts of the joint wherethe space between bones is wider or narrower.
"Using this technique, we'll hopefully be able to identify osteoarthritis earlier, and look at potential treatments before it becomes debilitating," said Turmezei,who is now a consultant at the Norfolk and NorwichUniversity Hospital's Department of Radiology. "Itcould be used to screen at-risk populations, such asthose with known arthritis, previous joint injury, orelite athletes who are at risk of developing arthritisdue to the continued strain placed on their joints."
While CT scanning is regularly used in the clinic todiagnose and monitor a range of health conditions,CT of joints has not yet been approved for use in research trials. According to the researchers, the success of the JSM algorithm demonstrates that 3Dimaging techniques have the potential to be more effective than 2D imaging. In addition, CT can now beused with very low doses of radiation, meaning thatit can be safely used more frequently for the purposesof ongoing monitoring.
"We've shown that this technique could be a valuabletool for the analysis of arthritis, in both clinical andresearch settings," said Turmezei. "When combinedwith 3D statistical analysis, it could also be used tospeed up the development of new treatments."