Shifting Focus: From Global Semantics to Local Prominent Features in Swin-Transformer for Knee Osteoarthritis Severity Assessment

Conventional imaging diagnostics frequently encounter bottle- necks due to manual inspection, which can lead to delays and inconsistencies. Although deep learning offers a pathway to au- tomation and enhanced accuracy, foundational models in com- puter vision often emphasize global context at the expense of lo- cal details, which are vital for medical imaging diagnostics. To address this, we harness the Swin Transformer’s capacity to dis- cern extended spatial dependencies within images through the hierarchical framework. Our novel contribution lies in refining local feature representations, orienting them specifically toward the final distribution of the classifier. This method ensures that local features are not only preserved but are also enriched with task-specific information, enhancing their relevance and detail at every hierarchical level. By implementing this strategy, our model demonstrates significant robustness and precision, as ev- idenced by extensive validation of two established benchmarks for Knee OsteoArthritis (KOA) grade classification. These re- sults highlight our approach’s effectiveness and its promising implications for the future of medical imaging diagnostics. Our implementation is available on Github.