Abstract
Femoroacetabular impingement (FAI) is caused by abnormal contact between the femur and acetabulum, resulting in pain, limited motion, and early osteoarthritis. Existing imaging techniques for diagnosing FAI face considerable challenges. Radiomics involves the quantitative extraction and analysis of imaging features using advanced algorithms, often combined with machine learning (ML), to enhance diagnostic and prognostic precision. When integrated with ML, radiomics can identify patterns beyond conventional imaging measurements, potentially enabling automated, precise, and reproducible assessment of hip morphology and pathology. Early studies demonstrate its potential to differentiate between normal, symptomatic, and asymptomatic cam-type hips. However, challenges persist, including the standardization of imaging protocols, feature selection, access to large datasets, and the explainability of models. This review summarizes the state of the art in radiomics for FAI and highlights its future applications.