Vendor-independent soft tissue lesion detection using weakly supervised and unsupervised adversarial domain adaptation

TL;DR

This paper presents a novel adversarial domain adaptation approach for vendor-independent soft tissue lesion detection in mammography, improving sensitivity across different device vendors with weakly supervised and unsupervised transfer learning.

Contribution

It introduces tailored transfer learning methods using adversarial learning that effectively adapt models across vendors with limited annotations, addressing domain shift in mammogram analysis.

Findings

Transfer learning improves detection sensitivity from 0.30 to 0.37 at 0.02 false positives per image.

Exam level annotations further increase sensitivity.

Proposed methods effectively handle class imbalance and domain differences.

Abstract

Computer-aided detection aims to improve breast cancer screening programs by helping radiologists to evaluate digital mammography (DM) exams. DM exams are generated by devices from different vendors, with diverse characteristics between and even within vendors. Physical properties of these devices and postprocessing of the images can greatly influence the resulting mammogram. This results in the fact that a deep learning model trained on data from one vendor cannot readily be applied to data from another vendor. This paper investigates the use of tailored transfer learning methods based on adversarial learning to tackle this problem. We consider a database of DM exams (mostly bilateral and two views) generated by Hologic and Siemens vendors. We analyze two transfer learning settings: 1) unsupervised transfer, where Hologic data with soft lesion annotation at pixel level and Siemens unlabelled data are used to annotate images in the latter data; 2) weak supervised transfer, where exam level labels for images from the Siemens mammograph are available. We propose tailored variants of recent state-of-the-art methods for transfer learning which take into account the class imbalance and incorporate knowledge provided by the annotations at exam level. Results of experiments indicate the beneficial effect of transfer learning in both transfer settings. Notably, at 0.02 false positives per image, we achieve a sensitivity of 0.37, compared to 0.30 of a baseline with no transfer. Results indicate that using exam level annotations gives an additional increase in sensitivity.