Publication: Objective stratification of knee osteoarthritis stages using a semi-supervised learning approach on multimodal MRI-CT cartilage features

Knee osteoarthritis is one of the most common joint diseases, affecting millions of middle-aged and older adults worldwide. Detecting the disease at an early stage is essential to slow its progression and improve treatment outcomes. However, current diagnosis often relies heavily on patients' pain perception, which does not always reflect the actual condition of the joint.

In this study, researchers developed a new artificial intelligence approach that combines MRI and CT imaging with semi-supervised machine learning to identify different stages of knee osteoarthritis. By using a limited number of expert-labelled scans alongside a larger dataset, the method was able to classify healthy, early-stage, and advanced cartilage degeneration with high reliability.

The analysis revealed clear structural changes in cartilage as the disease progresses, including thinning, surface irregularities, and increasing tissue heterogeneity. These findings demonstrate that AI-assisted image analysis can provide a more objective assessment of joint degeneration, paving the way for improved imaging biomarkers and future tools that support earlier diagnosis, more accurate disease monitoring, and better treatment decisions.

Read the full article here.