Sharma, from the Faculty of Science, is co-first author on a pioneering study that combines various biological markers from patient samples - such as blood and joint fluid - to identify distinct patient groups and predict their responses to joint replacement surgery.
Divya Sharma
The study was published in Annals of the Rheumatic Diseases and conducted in collaboration with Mohit Kapoor, Tier 1 Canada Research Chair in the Mechanisms of Joint Degeneration at the University of Toronto, and the research team at the Kapoor Lab, Schroeder Arthritis Institute and University Health Network (UHN).
This research is the first of its kind to comprehensively integrate and analyze multiple biological markers linked to knee osteoarthritis, a leading cause of disability worldwide.
The study, which examined approximately 2,700 features across patient samples, enables researchers to identify distinct biological subtypes - known as "endotypes" - among patients. Using clustering techniques, researchers were able to identify new endotypes of knee osteoarthritis and advance understanding of replacement surgery outcomes.
"Our research marks a significant leap forward," says Sharma. "Instead of relying solely on traditional clinical assessments, we can now offer a more refined and precise approach to understanding knee osteoarthritis, which is essential for improving patient outcomes."
The application of advanced deep learning techniques in this study allows for novel insights into how patients may respond to joint replacement surgery. This data-driven methodology has the potential to reshape treatment plans, offering tailored therapeutic strategies that move beyond a one-size-fits-all model.
Clinicians often base surgical recommendations on a limited subset of patient data, which, according to Sharma, can contribute to variability and unpredictability in outcomes.
This study changes that landscape by effectively combining clinical data with extensive biological information, thereby enhancing the predictive capabilities surrounding surgical success.
"By utilizing the power of AI and machine learning, we are moving towards a more personalized treatment paradigm for osteoarthritis," says Sharma. "It's about understanding the unique biological makeup of each patient to provide the most effective care."
The implications of this research extend beyond academia, she adds. The study has captured the attention of the medical community and was featured in UHN News.
This story was originally featured in YFile, York University's community newsletter.
