# Generative augmentations for improved cardiac ultrasound segmentation using diffusion models

## TL;DR

This paper shows how using generative augmentations with diffusion models improves the accuracy of cardiac ultrasound segmentation models without needing more labeled data.

## Contribution

The novelty is using diffusion models for generative augmentations to boost segmentation model generalizability in cardiac ultrasound.

## Key findings

- Generative augmentations improved segmentation robustness by over 20mm in Hausdorff distance on external datasets.
- Automatic ejection fraction estimation limits of agreement improved by up to 20% on out-of-distribution cases.
- Experts could not distinguish between real and generated images, showing high quality of augmentations.

## 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 generative 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 (EF) estimation improved by up to 20% of absolute EF 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.