Pre and Post-hoc Diagnosis and Interpretation of Malignancy from Breast DCE-MRI

Abstract

We propose a new method for breast cancer screening from DCE-MRI basedon a post-hoc approach that is trained using weakly annotated data (i.e.,labels are available only at the image level without any lesion delineation).Our proposed post-hoc method automatically diagnosis the whole volumeand, for positive cases, it localizes the malignant lesions that led to suchdiagnosis. Conversely, traditional approaches follow a pre-hoc approach thatinitially localises suspicious areas that are subsequently classified to establishthe breast malignancy – this approach is trained using strongly annotateddata (i.e., it needs a delineation and classification of all lesions in an image). We also aim to establish the advantages and disadvantages of bothapproaches when applied to breast screening from DCE-MRI. Relying on experiments on a breast DCE-MRI dataset that contains scans of 117 patients,our results show that the post-hoc method is more accurate for diagnosingthe whole volume per patient, achieving an AUC of 0.91, while the pre-hoc method achieves an AUC of 0.81. However, the performance for localisingthe malignant lesions remains challenging for the post-hoc method due tothe weakly labelled dataset employed during training.

Publication
Preprint
Date
Links