Feature-Conditioned Cascaded Video Diffusion Models for Precise Echocardiogram Synthesis

TL;DR

This paper introduces a novel diffusion-based method to generate realistic echocardiogram videos from single images and clinical parameters, significantly improving the accuracy of clinical metric predictions.

Contribution

It extends diffusion models to generate video sequences from single images conditioned on clinical data, enabling retrospective analysis and training data augmentation in echocardiography.

Findings

Achieved 93% R^2 score in predicting Left Ventricle Ejection Fraction.

Outperformed recent sequence-to-sequence methods by 38 points in R^2 score.

Demonstrated the feasibility of generating plausible echocardiogram videos from minimal input.

Abstract

Image synthesis is expected to provide value for the translation of machine learning methods into clinical practice. Fundamental problems like model robustness, domain transfer, causal modelling, and operator training become approachable through synthetic data. Especially, heavily operator-dependant modalities like Ultrasound imaging require robust frameworks for image and video generation. So far, video generation has only been possible by providing input data that is as rich as the output data, e.g., image sequence plus conditioning in, video out. However, clinical documentation is usually scarce and only single images are reported and stored, thus retrospective patient-specific analysis or the generation of rich training data becomes impossible with current approaches. In this paper, we extend elucidated diffusion models for video modelling to generate plausible video sequences from single images and arbitrary conditioning with clinical parameters. We explore this idea within the context of echocardiograms by looking into the variation of the Left Ventricle Ejection Fraction, the most essential clinical metric gained from these examinations. We use the publicly available EchoNet-Dynamic dataset for all our experiments. Our image to sequence approach achieves an $R^2$ score of 93%, which is 38 points higher than recently proposed sequence to sequence generation methods. Code and models will be available at: https://github.com/HReynaud/EchoDiffusion.