Lesion-Aware Generative Artificial Intelligence for Virtual Contrast-Enhanced Mammography in Breast Cancer

TL;DR

This paper introduces Seg-CycleGAN, a lesion-aware deep learning model that synthesizes high-quality virtual contrast-enhanced mammography images from low-energy scans, aiming to reduce radiation and contrast risks.

Contribution

The study presents a novel lesion-guided generative framework that improves virtual contrast image synthesis by incorporating lesion segmentation into the CycleGAN architecture.

Findings

Seg-CycleGAN outperforms baseline in PSNR and SSIM metrics.

Lesion fidelity is significantly improved in generated images.

Model maintains competitive MSE and VIF scores.

Abstract

Contrast-Enhanced Spectral Mammography (CESM) is a dual-energy mammographic technique that improves lesion visibility through the administration of an iodinated contrast agent. It acquires both a low-energy image, comparable to standard mammography, and a high-energy image, which are then combined to produce a dual-energy subtracted image highlighting lesion contrast enhancement. While CESM offers superior diagnostic accuracy compared to standard mammography, its use entails higher radiation exposure and potential side effects associated with the contrast medium. To address these limitations, we propose Seg-CycleGAN, a generative deep learning framework for Virtual Contrast Enhancement in CESM. The model synthesizes high-fidelity dual-energy subtracted images from low-energy images, leveraging lesion segmentation maps to guide the generative process and improve lesion reconstruction. Building upon the standard CycleGAN architecture, Seg-CycleGAN introduces localized loss terms focused on lesion areas, enhancing the synthesis of diagnostically relevant regions. Experiments on the CESM@UCBM dataset demonstrate that Seg-CycleGAN outperforms the baseline in terms of PSNR and SSIM, while maintaining competitive MSE and VIF. Qualitative evaluations further confirm improved lesion fidelity in the generated images. These results suggest that segmentation-aware generative models offer a viable pathway toward contrast-free CESM alternatives.