South Korean researchers have developed a new test that diagnoses arthritis in 10 minutes and identifies its type and severity with over 98 percent accuracy.
The test uses artificial intelligence and a specialized gold nanoparticle sensor to analyze synovial fluid—the lubricating liquid found in joints—offering a much faster and more accurate diagnosis than traditional methods.
The diagnostic platform, developed by the Korea Institute of Materials Science in partnership with Seoul St. Mary’s Hospital, is capable of distinguishing between osteoarthritis and rheumatoid arthritis, two conditions that often present with similar symptoms such as joint pain, swelling, and stiffness.
While current diagnostic methods rely on a combination of imaging, lab tests, and clinical evaluations that can take weeks, this new test offers a rapid, cost-effective alternative that requires only a small fluid sample and minimal equipment.
The technology works by using surface-enhanced Raman scattering (SERS), a method that detects molecular vibrations to produce a chemical fingerprint of the fluid. The researchers designed a paper-based sensor coated with sea urchin-shaped gold nanoparticles that enhances the signal strength, allowing for more precise readings. Once the spectral data is collected, it is processed using a machine learning algorithm known as a support vector machine.
In clinical testing involving 120 patients, the system achieved up to 98.1 percent diagnostic accuracy, with a sensitivity rate of 97.3 percent and a specificity rate of 100 percent when distinguishing between arthritis types.
Beyond just diagnosing the type of arthritis, the platform also has the ability to determine the severity of rheumatoid arthritis. By analyzing the concentration of white blood cells in synovial fluid, the system grouped patient samples by disease severity and reached accuracy levels above 98 percent in identifying which group a patient belonged to.
The researchers used additional mathematical tools such as non-negative matrix factorization and Pearson correlation coefficients to link molecular patterns to clinical symptoms, helping uncover distinct metabolic signatures associated with each condition.
Traditional diagnostics for arthritis, such as MRIs, X-rays, and blood tests, are not only time-consuming and expensive but often fail to detect early-stage disease.
This new method provides an alternative that is not only faster but more precise, especially in the early phases when treatment is most effective. The sensor is manufactured through a simple process and does not require complex lab equipment, making it suitable for use in regular clinical settings and potentially more accessible to patients.
The study’s authors noted that synovial fluid contains over 1,000 different metabolites, offering a clearer view of joint health compared to blood-based tests. They believe this breakthrough could serve as a valuable pre-screening tool before doctors move forward with more costly imaging.
While the current study involved a limited sample size, further research is already underway to expand the data and explore how the technology can be used to track disease progression over time.
Researchers involved in the project say the platform’s use could eventually expand beyond arthritis to diagnose other diseases that affect bodily fluids. With the possibility of earlier detection and more personalized treatment plans, this development offers hope to the millions worldwide suffering from chronic joint pain and inflammation.