Accurate and Efficient Fetal Birth Weight Estimation from 3D Ultrasound

2025-06-30

AI Summary

Existing fetal birth weight (FBW) estimation methods suffer from operator dependency, limited spatial information in 2D ultrasound, and suboptimal accuracy. To address these limitations, this paper proposes the first end-to-end FBW estimation framework based on 3D fetal ultrasound volumetric data. Methodologically, we design a Multi-Scale Feature Fusion Network (MFFN) incorporating joint channel-spatial attention, and introduce a Semi-Supervised Synthetic Learning Framework (SSLF) to mitigate label scarcity. Furthermore, we innovatively adopt an ordinal loss function to enhance regression performance under sparse supervision. Experimental results demonstrate a mean absolute error of 166.4 ± 155.9 g and a mean absolute percentage error of 5.1 ± 4.6%, significantly outperforming state-of-the-art approaches and achieving diagnostic accuracy comparable to experienced clinicians. The source code is publicly available.

Technology Category

• Machine Learning: Multimodal Learning
• Intelligent Robots: State Estimation
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Accurate fetal birth weight (FBW) estimation is essential for optimizing delivery decisions and reducing perinatal mortality. However, clinical methods for FBW estimation are inefficient, operator-dependent, and challenging to apply in cases of complex fetal anatomy. Existing deep learning methods are based on 2D standard ultrasound (US) images or videos that lack spatial information, limiting their prediction accuracy. In this study, we propose the first method for directly estimating FBW from 3D fetal US volumes. Our approach integrates a multi-scale feature fusion network (MFFN) and a synthetic sample-based learning framework (SSLF). The MFFN effectively extracts and fuses multi-scale features under sparse supervision by incorporating channel attention, spatial attention, and a ranking-based loss function. SSLF generates synthetic samples by simply combining fetal head and abdomen data from different fetuses, utilizing semi-supervised learning to improve prediction performance. Experimental results demonstrate that our method achieves superior performance, with a mean absolute error of $166.4pm155.9$ $g$ and a mean absolute percentage error of $5.1pm4.6$%, outperforming existing methods and approaching the accuracy of a senior doctor. Code is available at: https://github.com/Qioy-i/EFW.

Problem

• Improves fetal birth weight estimation accuracy using 3D ultrasound
• Addresses inefficiency and operator-dependency in current clinical methods
• Overcomes spatial information limitations in existing deep learning approaches

Innovation

• Direct FBW estimation from 3D ultrasound volumes
• Multi-scale feature fusion with attention mechanisms
• Synthetic sample learning for improved accuracy