Ultrasound Image Synthesis Using Generative AI for Lung Ultrasound Detection

TL;DR

This paper introduces DiffUltra, a novel generative AI method that synthesizes diverse and realistic lung ultrasound images, enhancing detection accuracy especially for rare cases and addressing data imbalance issues in healthcare AI models.

Contribution

DiffUltra is the first generative AI technique to produce realistic lung ultrasound images with lesion variability, guided by a Lesion-anatomy Bank, improving detection performance.

Findings

Improves consolidation detection by 5.6% AP.

Increases rare case detection by 25% AP.

Enhances data diversity and model robustness.

Abstract

Developing reliable healthcare AI models requires training with representative and diverse data. In imbalanced datasets, model performance tends to plateau on the more prevalent classes while remaining low on less common cases. To overcome this limitation, we propose DiffUltra, the first generative AI technique capable of synthesizing realistic Lung Ultrasound (LUS) images with extensive lesion variability. Specifically, we condition the generative AI by the introduced Lesion-anatomy Bank, which captures the lesion's structural and positional properties from real patient data to guide the image synthesis.We demonstrate that DiffUltra improves consolidation detection by 5.6% in AP compared to the models trained solely on real patient data. More importantly, DiffUltra increases data diversity and prevalence of rare cases, leading to a 25% AP improvement in detecting rare instances such as large lung consolidations, which make up only 10% of the dataset.