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

TL;DR

This paper introduces a post-hoc method for breast cancer diagnosis from DCE-MRI that uses weakly annotated data, demonstrating higher accuracy in overall diagnosis but challenges in lesion localization compared to traditional pre-hoc methods.

Contribution

It presents a novel post-hoc approach trained with weak labels, contrasting with traditional pre-hoc methods requiring detailed annotations, and evaluates their relative advantages in breast cancer screening.

Findings

Post-hoc method achieves 0.91 AUC for diagnosis.

Pre-hoc method achieves 0.81 AUC for diagnosis.

Lesion localization remains challenging for the post-hoc approach.

Abstract

We propose a new method for breast cancer screening from DCE-MRI based on 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 volume and, for positive cases, it localizes the malignant lesions that led to such diagnosis. Conversely, traditional approaches follow a pre-hoc approach that initially localises suspicious areas that are subsequently classified to establish the breast malignancy -- this approach is trained using strongly annotated data (i.e., it needs a delineation and classification of all lesions in an image). Another goal of this paper is to establish the advantages and disadvantages of both approaches 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 diagnosing the 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 localising the malignant lesions remains challenging for the post-hoc method due to the weakly labelled dataset employed during training.