Joint Loading Patterns and Knee Osteoarthritis Progression

Date(s) - 10/17/2022
3:00 pm - 4:00 pm

Communicore, C1-009

Kerry Costello, Ph.D., Assistant Professor, Department of Mechanical & Aerospace Engineering, University of Florida

Dr. Kerry Costello is an assistant professor in the department of Mechanical and Aerospace Engineering. Prior to joining the University of Florida, she completed a postdoctoral fellowship at Boston University in the Department of Physical Therapy & Athletic Training and the Section of Rheumatology; a doctoral degree in biomedical engineering at Dalhousie University in Halifax, Nova Scotia, Canada; a master’s degree in biomedical engineering at Virginia Tech; and her undergraduate degree in biomedical and mechanical engineering at Duke University. Dr. Costello also spent a year doing research at a private orthopedic sports medicine research institute in Colorado and a year completing a Fulbright scholarship at Vrije Universiteit Amsterdam, the Netherlands. Her research utilizes motion capture data, wearable sensor data, and signal analysis and machine learning tools to understand how time-varying, multidimensional joint loading patterns during human movement contribute to disease progression in knee osteoarthritis. She also created, hosts, and produces the Osteoarthritis Research Society International’s ‘Hey OA’ podcast.


Mechanical loading on the knee joint during human movement is one of the only modifiable risk factors for knee osteoarthritis, a painful disease affecting over 350 million people worldwide. Gait analysis studies have identified key features of joint loading during walking that are associated with disease progression, in particular the knee adduction moment magnitude. However, the tissues of the joint respond not only to the magnitude, but also to the time-varying, multi-dimensional patterns of joint loading exposure. Better characterization of these loading patterns and their role in the disease process could lead to improved conservative management for knee osteoarthritis, such as patient-specific recommendations for biomechanical interventions, physical therapy, and/or physical activity type, intensity, and frequency. The technologies used to capture gait data in a laboratory setting and physical activity data in a real-world setting (e.g., accelerometers) provide a wealth of detailed information about how people move and the associated loading patterns during movement. However, the complex interactions among gait, physical activity, and patient-specific factors (e.g., age, sex, disease severity) and the time-varying, multidimensional nature of these signals make traditional analyses challenging. This talk presents research exploring data science and machine learning approaches to analyze these complex loading patterns and provide insight into the role of human movement in knee osteoarthritis progression.