Development of an Artificial Intelligence Model to Predict the Progression Speed and Prognosis of Knee Osteoarthritis

Professor Lee Yong-seok's Team at Bundang Seoul National University Hospital
Utilizes Machine Learning Techniques... Prognosis Prediction Accuracy 88%

Professor Lee Yong-seok, Department of Orthopedic Surgery, Bundang Seoul National University Hospital.

[Asia Economy Reporter Lee Gwan-joo] A research team led by Professor Lee Yong-seok of the Department of Orthopedics at Bundang Seoul National University Hospital announced on the 14th that they have developed a model using artificial intelligence (AI) to predict the progression speed and prognosis of knee osteoarthritis.

The team applied machine learning, one of the AI techniques that allows computers to find models based on data without specific instructions. For this, they utilized demographic data (age, gender, body mass index, bone density, etc.), occupation (jobs that may affect the knee), comorbidities (hypertension, diabetes, dementia, etc.), and radiological data (initial knee information, surgical information, etc.) from 83,280 osteoarthritis patients who visited between 2003 and 2019.

The predictive model developed by the team showed accuracies of 71% and 88% in predicting the progression speed and prognosis of osteoarthritis, respectively. The most significant factors affecting osteoarthritis were the initial severity of the condition and occupations impacting the knee. Utilizing this model is expected to help diagnose patient conditions and determine future treatment and prognosis. In particular, since it can analyze and predict progression based on individual patient conditions without requiring complex procedures for diagnosing osteoarthritis, it is anticipated to be easily used even in primary care clinics.

Professor Lee explained, “Consistent management is necessary to treat osteoarthritis, but existing methods had various issues such as effectiveness and cost. Using this model, it will be convenient and relatively inexpensive to predict treatment methods and prognosis even in primary care settings.”

The study was published in the SCI-level journal, Journal of Orthopaedic Research.

This content was produced with the assistance of AI translation services.