Attention-Gated Networks for Improving Ultrasound Scan Plane Detection

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.

Submitted to MIDL2018 (OpenReview: https://openreview.net/forum?id=BJtn7-3sM)

Cited by in corpus (14)

• RA-UNet: A hybrid deep attention-aware network to extract liver and tumor in CT scans
• SLSNet: Skin lesion segmentation using a lightweight generative adversarial network
• Ultrasound Image Representation Learning by Modeling Sonographer Visual Attention
• WearMask: Fast In-browser Face Mask Detection with Serverless Edge Computing for COVID-19
• Efficient Scopeformer: Towards Scalable and Rich Feature Extraction for Intracranial Hemorrhage Detection
• PDANet: Pyramid Density-aware Attention Net for Accurate Crowd Counting
• Adaptive Early-Learning Correction for Segmentation from Noisy Annotations
• 3D Grid-Attention Networks for Interpretable Age and Alzheimer's Disease Prediction from Structural MRI
• SAR-U-Net: squeeze-and-excitation block and atrous spatial pyramid pooling based residual U-Net for automatic liver segmentation in Computed Tomography
• CNNs, LSTMs, and Attention Networks for Pathology Detection in Medical Data
• Algorithm Fairness in AI for Medicine and Healthcare
• AI-Based Detection, Classification and Prediction/Prognosis in Medical Imaging: Towards Radiophenomics
• Tissue characterization based on the analysis on i3DUS data for diagnosis support in neurosurgery
• E(2) Equivariant Self-Attention for Radio Astronomy