
This study developed machine learning models incorporating radiomics features to differentiate between benign and malignant adnexal masses with solid morphology. A total of 68 radiomics features were analyzed from ultrasound images of 326 patients, leading to models with moderate discriminative ability (AUC: 0.80). Comparison with the ADNEX model and subjective assessments showed that while radiomics models had lower sensitivity and specificity, their potential warrants further development.
Like
Save
Share