S. Degadwala

This review categorizes the methods used for grading KOA into handcrafted/radiomics, convolutional and transfer-learning systems, ordinal or hi-erarchical models, transformer and vision–language and fusion-oriented pipe-lines, and priority directions for clinically credible de-ployment include clinically relevant measures of uncer-tainty, quantitative assessment of the quality of explanations, and imaging integrated with clinical or longitudinal evidence.

The findings highlight the need for robust AI models that can accurately assess disease severity and provide meaningful explanations across different agricultural settings and suggest future research directions toward the goal of building reliable, interpretable, and practical systems for precision agriculture.