Developing image analysis methods to evaluate cartilage degeneration in animal models of osteoarthritis

Osteoarthritis (OA) is a common joint disease which can cause cartilage loss, joint damage, and pain. An important focus area for OA researchers is the cartilage degeneration that happens at early stages of the disease. A better understanding of these early changes can help researchers find more effective early treatments for OA. To do this, researchers often use microscopic-level imaging methods to observe changes in cartilage and cartilage cells that are related to cartilage degeneration. To understand and to measure those changes, researchers need to delineate the cartilage region and the cartilage cells from the microscopic-level images which often include neighboring tissue types such as bone. This process allows investigators to put numbers to different features including cartilage thickness, number and distribution of cartilage cells, and level of cell function. Traditionally, finding edges of cartilage or cells involves researchers using a mouse and drawing the cartilage/cell boundaries on the computer screen. However, this approach is very slow, laborious, and prone to error, which can significantly influence the quality of the information that is collected and slow the progress of OA research.

This thesis work aimed to address the shortcomings of manual identification by developing automated, computer-based, image analysis algorithms to delineate/identify cartilage and cartilage cells from microscopic-level images saved as computer files. The accuracy of each algorithm that was developed was checked to ensure the automated methods were achieving results that were as accurate as those from expert OA researchers. These automated algorithms were then used in several example studies to rapidly and accurately measure the differences in cartilage and cell features between injured and healthy cartilage. It is expected that the algorithms developed in this work will help OA researchers better and more quickly understand tissue- and cell-level changes in the cartilage that occur during the development of OA.