Generative augmentations for improved cardiac ultrasound segmentation using diffusion models

2025-02-27

AI Summary

Cardiac ultrasound segmentation faces challenges including scarcity of annotated data, poor cross-center generalizability, and inconsistent annotation standards. This work introduces, for the first time, denoising diffusion probabilistic models (DDPMs) for generative data augmentation of cardiac ultrasound images, producing high-fidelity synthetic images indistinguishable from real ones in expert blinded evaluations. Our method significantly enhances model robustness across domains using augmentation alone: on multi-center testing, the Hausdorff distance decreases by over 20 mm, and the limits of agreement in Bland–Altman analysis for ejection fraction estimation improve by 20 percentage points. This study advances the trustworthy deployment of generative augmentation in medical ultrasound, and we publicly release the open-source toolkit EchoGAINS to support reproducible research and clinical translation.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

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: Trust and reliance of crowd workers and data experts on GenAI
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

One of the main challenges in current research on segmentation in cardiac ultrasound is the lack of large and varied labeled datasets and the differences in annotation conventions between datasets. This makes it difficult to design robust segmentation models that generalize well to external datasets. This work utilizes diffusion models to create generative augmentations that can significantly improve diversity of the dataset and thus the generalisability of segmentation models without the need for more annotated data. The augmentations are applied in addition to regular augmentations. A visual test survey showed that experts cannot clearly distinguish between real and fully generated images. Using the proposed generative augmentations, segmentation robustness was increased when training on an internal dataset and testing on an external dataset with an improvement of over 20 millimeters in Hausdorff distance. Additionally, the limits of agreement for automatic ejection fraction estimation improved by up to 20% of absolute ejection fraction value on out of distribution cases. These improvements come exclusively from the increased variation of the training data using the generative augmentations, without modifying the underlying machine learning model. The augmentation tool is available as an open source Python library at https://github.com/GillesVanDeVyver/EchoGAINS.

Problem

• Enhances cardiac ultrasound segmentation robustness
• Utilizes diffusion models for dataset diversity
• Improves model generalizability without extra annotations

Innovation

• Uses diffusion models for augmentation
• Enhances dataset diversity without annotation
• Improves segmentation robustness significantly