Diffusion Model-based Data Augmentation Method for Fetal Head Ultrasound Segmentation

2025-06-30

AI Summary

To address the scarcity, high annotation cost, and privacy constraints limiting labeled fetal cranial ultrasound data, this paper proposes a mask-guided diffusion model that jointly synthesizes high-fidelity ultrasound images and their corresponding segmentation masks—enabling paired synthetic data augmentation. Crucially, the method embeds segmentation priors directly into the diffusion process, substantially improving the anatomical plausibility and task relevance of generated samples. These synthetic data are subsequently employed to supervise fine-tuning of the Segment Anything Model (SAM). Evaluated on two real-world low-resource cohorts—Spanish and African—the approach achieves state-of-the-art Dice scores of 94.66% and 94.38%, respectively. Results demonstrate both the efficacy and strong cross-domain generalizability of synthetic data for medical image segmentation under severe data scarcity.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Security and Privacy: Data transparency and provenance

Abstract

Medical image data is less accessible than in other domains due to privacy and regulatory constraints. In addition, labeling requires costly, time-intensive manual image annotation by clinical experts. To overcome these challenges, synthetic medical data generation offers a promising solution. Generative AI (GenAI), employing generative deep learning models, has proven effective at producing realistic synthetic images. This study proposes a novel mask-guided GenAI approach using diffusion models to generate synthetic fetal head ultrasound images paired with segmentation masks. These synthetic pairs augment real datasets for supervised fine-tuning of the Segment Anything Model (SAM). Our results show that the synthetic data captures real image features effectively, and this approach reaches state-of-the-art fetal head segmentation, especially when trained with a limited number of real image-mask pairs. In particular, the segmentation reaches Dice Scores of 94.66% and 94.38% using a handful of ultrasound images from the Spanish and African cohorts, respectively. Our code, models, and data are available on GitHub.

Problem

• Limited access to medical image data due to privacy constraints
• High cost and time for manual annotation by experts
• Need for synthetic data to improve fetal ultrasound segmentation

Innovation

• Diffusion models generate synthetic ultrasound images
• Mask-guided approach pairs images with segmentation
• Augments real data for SAM fine-tuning