Hybrid Diffusion Model for Breast Ultrasound Image Augmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Abir, Farhan Fuad, Jennifer, Sanjeda Sara, Yousefi, Niloofar, Brattain, Laura J.
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911548890415104
author Abir, Farhan Fuad
Jennifer, Sanjeda Sara
Yousefi, Niloofar
Brattain, Laura J.
author_facet Abir, Farhan Fuad
Jennifer, Sanjeda Sara
Yousefi, Niloofar
Brattain, Laura J.
contents We propose a hybrid diffusion-based augmentation framework to overcome the critical challenge of ultrasound data augmentation in breast ultrasound (BUS) datasets. Unlike conventional diffusion-based augmentations, our approach improves visual fidelity and preserves ultrasound texture by combining text-to-image generation with image-to-image (img2img) refinement, as well as fine-tuning with low-rank adaptation (LoRA) and textual inversion (TI). Our method generated realistic, class-consistent images on an open-source Kaggle breast ultrasound image dataset (BUSI). Compared to the Stable Diffusion v1.5 baseline, incorporating TI and img2img refinement reduced the Frechet Inception Distance (FID) from 45.97 to 33.29, demonstrating a substantial gain in fidelity while maintaining comparable downstream classification performance. Overall, the proposed framework effectively mitigates the low-fidelity limitations of synthetic ultrasound images and enhances the quality of augmentation for robust diagnostic modeling.
format Preprint
id arxiv_https___arxiv_org_abs_2603_26834
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Hybrid Diffusion Model for Breast Ultrasound Image Augmentation
Abir, Farhan Fuad
Jennifer, Sanjeda Sara
Yousefi, Niloofar
Brattain, Laura J.
Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
We propose a hybrid diffusion-based augmentation framework to overcome the critical challenge of ultrasound data augmentation in breast ultrasound (BUS) datasets. Unlike conventional diffusion-based augmentations, our approach improves visual fidelity and preserves ultrasound texture by combining text-to-image generation with image-to-image (img2img) refinement, as well as fine-tuning with low-rank adaptation (LoRA) and textual inversion (TI). Our method generated realistic, class-consistent images on an open-source Kaggle breast ultrasound image dataset (BUSI). Compared to the Stable Diffusion v1.5 baseline, incorporating TI and img2img refinement reduced the Frechet Inception Distance (FID) from 45.97 to 33.29, demonstrating a substantial gain in fidelity while maintaining comparable downstream classification performance. Overall, the proposed framework effectively mitigates the low-fidelity limitations of synthetic ultrasound images and enhances the quality of augmentation for robust diagnostic modeling.
title Hybrid Diffusion Model for Breast Ultrasound Image Augmentation
topic Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.26834