Computed Tomography Radiogenomics: Predicting Molecular Subtypes in Gastric Stromal Tumors

A combined model integrating clinical and contrast-enhanced computed tomography (CE-CT) features accurately predicts specific genetic mutations in gastrointestinal stromal tumors (GISTs). Researchers developed three models from data from 231 GIST patients, with the best model achieving high accuracy in predicting molecular subtypes. In the training cohort, the area under the curve (AUC) values indicated significant accuracy, suggesting that preoperative CE-CT radiogenomics may guide personalized precision therapy in GIST patients.