Researchers at the Korea Institute of Materials Science (KIMS), in partnership with Seoul St. Mary’s Hospital, have developed a plasmonic sensor capable of diagnosing osteoarthritis (OA) and rheumatoid arthritis (RA) in just 10 minutes using synovial fluid (SVF). The tool distinguishes between the two diseases and assesses the severity of rheumatoid arthritis with over 98% accuracy. Details of this new diagnostic tool are published in Advanced Healthcare Materials.
The new diagnostic platform improves upon traditional methods such as X-rays, MRI, and blood tests, which are time-consuming, costly, and often inconclusive in early disease stages. “Despite differences in pathophysiology, these diseases share overlapping features that complicate diagnosis, necessitating early, more accurate, and cost-effective diagnostic tools,” the researchers wrote.
For the new tool, the team employed surface-enhanced Raman scattering (SERS) and developed a diagnostic method that uses a sensor composed of a sea urchin-shaped gold nanostructure on paper with high moisture absorption. Data from the SERS analysis leverages AI and are processed using a support vector machine (SVM)-based machine learning algorithm which can differentiate disease types and severity levels based on trace levels of biomarkers.
In a clinical study involving 120 patients, the tool achieved higher than 94% accuracy in distinguishing between osteoarthritis and rheumatoid arthritis and more than 95% accuracy in assessing rheumatoid arthritis severity. The team accomplished this by evaluating the Raman spectra of synovial fluid samples and then correlating these data with hematology results and known metabolic biomarkers using non-negative matrix factorization (NMF) and Pearson correlation coefficients (PCC).
This research builds on earlier investigations into the metabolomic differences in synovial fluid in arthritis patients. Synovial fluid, which bathes the joints, contains more than 1,000 different metabolites and reflects joint health more directly than blood tests. Past diagnostic attempts have used methods such as liquid chromatography-mass spectrometry (LC-MS) and nuclear magnetic resonance (NMR), which can effectively diagnose joint health, but are costly and labor-intensive. “Although metabolic biomarkers for OA and RA in SVF have not been fully discovered, the metabolic composition of SVF changes significantly in inflammatory diseases such as RA and degenerative diseases such as OA, making it an invaluable source for identifying disease-specific biomarkers,” the researchers wrote.
The team’s approach leverages the benefits of SERS, a technique capable of detecting minute molecular signals, in combination with mathematical algorithms to enhance signal interpretation and accuracy. “The identification of metabolites showing significant correlations with Raman spectral features associated with arthritis is expected to provide valuable insights into biomarker discovery, thereby enhancing the potential for early and accurate diagnosis of arthritis,” the researchers wrote.
The researchers noted that the potential benefits of this new diagnostic method lie in its simplicity, speed, and accuracy. The method requires minimal sample preparation and can be administered in clinical settings without advanced imaging equipment.
The developers noted that while the 120-patient sample size is significant, it may not fully account for inter-patient variability in broader populations. Further, the Raman spectral library of metabolites used for classification, though advanced, is still developing. Future studies will look to further validate these findings via longitudinal studies to assess how well the platform monitors disease progression over time.
The researchers also noted that this approach may be applicable beyond arthritis diagnosis. “We also plan to continue expanding our research to cover a wider range of diseases in the future,” Jung said.
