Deep learning radiomics model bests machine learning for parotid tumor classification on MRI

A deep learning radiomics model (TabResNet) on fat-suppressed T2-weighted MRI achieved AUC 0.8913 for differentiating benign vs malignant parotid tumors, outperforming Random Forest (AUC 0.8641) in 102 patients.

• The study used an independent held-out test set, with five-fold cross-validation only for regularization parameter selection during feature screening.
• A multi-reader trial showed that TabResNet model assistance improved diagnostic performance and reading efficiency for clinicians of varying seniority.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.