Attention-Gated Networks for Improving Ultrasound Scan Plane Detection

TL;DR

This paper introduces an attention-gated network for real-time fetal ultrasound scan plane detection, enhancing localization accuracy and interpretability, especially in low-quality images, by incorporating a trainable soft-attention mechanism.

Contribution

It proposes a generic, end-to-end trainable attention mechanism that improves ultrasound scan plane detection and interpretability, adaptable to various architectures with minimal additional parameters.

Findings

Improves object localization in ultrasound images.

Reduces false positives significantly.

Enhances model interpretability through attention maps.

Abstract

In this work, we apply an attention-gated network to real-time automated scan plane detection for fetal ultrasound screening. Scan plane detection in fetal ultrasound is a challenging problem due the poor image quality resulting in low interpretability for both clinicians and automated algorithms. To solve this, we propose incorporating self-gated soft-attention mechanisms. A soft-attention mechanism generates a gating signal that is end-to-end trainable, which allows the network to contextualise local information useful for prediction. The proposed attention mechanism is generic and it can be easily incorporated into any existing classification architectures, while only requiring a few additional parameters. We show that, when the base network has a high capacity, the incorporated attention mechanism can provide efficient object localisation while improving the overall performance. When the base network has a low capacity, the method greatly outperforms the baseline approach and significantly reduces false positives. Lastly, the generated attention maps allow us to understand the model's reasoning process, which can also be used for weakly supervised object localisation.