Automatic diagnosis of knee osteoarthritis severity using Swin transformer

TL;DR

This paper presents an automated method using Swin Transformer and multi-prediction architecture to accurately diagnose and assess the severity of knee osteoarthritis from radiographic images, aiding early clinical intervention.

Contribution

It introduces a novel multi-prediction head architecture and a training approach to improve generalization across datasets for KOA severity prediction.

Findings

High accuracy in KOA severity prediction

Effective reduction of data drift between datasets

Demonstrated feasibility of Swin Transformer in medical imaging

Abstract

Knee osteoarthritis (KOA) is a widespread condition that can cause chronic pain and stiffness in the knee joint. Early detection and diagnosis are crucial for successful clinical intervention and management to prevent severe complications, such as loss of mobility. In this paper, we propose an automated approach that employs the Swin Transformer to predict the severity of KOA. Our model uses publicly available radiographic datasets with Kellgren and Lawrence scores to enable early detection and severity assessment. To improve the accuracy of our model, we employ a multi-prediction head architecture that utilizes multi-layer perceptron classifiers. Additionally, we introduce a novel training approach that reduces the data drift between multiple datasets to ensure the generalization ability of the model. The results of our experiments demonstrate the effectiveness and feasibility of our approach in predicting KOA severity accurately.