FPDANet: A Multi-Section Classification Model for Intelligent Screening of Fetal Ultrasound

2025-06-06

AI Summary

Fetal ultrasound images pose significant classification challenges due to low contrast, high noise, and substantial inter-class similarity. To address these issues, this paper proposes a dual-path multi-scale network built upon ResNet. It incorporates a Depth-aware Position Attention Mechanism (DAN) to model spatial dependencies and introduces a Bilateral Multi-scale Information Fusion Module (FPAN) to enable bidirectional cooperative enhancement between cross-scale contextual features and global representations. This architecture is the first to integrate position-aware attention with bidirectional multi-scale feature interaction, markedly improving fine-grained discriminative capability. Evaluated on a multi-class fetal ultrasound classification task, the method achieves 91.05% Top-1 accuracy and 100% Top-5 accuracy—outperforming state-of-the-art models. The approach offers a clinically viable, interpretable, and robust solution for intelligent prenatal screening.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

ResNet has been widely used in image classification tasks due to its ability to model the residual dependence of constant mappings for linear computation. However, the ResNet method adopts a unidirectional transfer of features and lacks an effective method to correlate contextual information, which is not effective in classifying fetal ultrasound images in the classification task, and fetal ultrasound images have problems such as low contrast, high similarity, and high noise. Therefore, we propose a bilateral multi-scale information fusion network-based FPDANet to address the above challenges. Specifically, we design the positional attention mechanism (DAN) module, which utilizes the similarity of features to establish the dependency of different spatial positional features and enhance the feature representation. In addition, we design a bilateral multi-scale (FPAN) information fusion module to capture contextual and global feature dependencies at different feature scales, thereby further improving the model representation. FPDANet classification results obtained 91.05% and 100% in Top-1 and Top-5 metrics, respectively, and the experimental results proved the effectiveness and robustness of FPDANet.

Problem

• Improves fetal ultrasound image classification accuracy
• Addresses low contrast and high noise in ultrasound images
• Enhances feature representation with attention mechanisms

Innovation

• Bilateral multi-scale information fusion network
• Positional attention mechanism (DAN) module
• Captures contextual and global feature dependencies