Robust Fetal Pose Estimation across Gestational Ages via Cross-Population Augmentation

2025-09-15

AI Summary

Existing fetal motion tracking methods rely on late-gestational 3D EPI time-series with manual annotations, limiting generalizability to early gestation due to dramatic anatomical changes and severe scarcity of early-stage labeled data. To address this, we propose a cross-population data augmentation framework featuring fetus-specific in-utero environment simulation and fetal pose distribution modeling—enabling robust generalization from late- to early-gestational data without early annotations. Our method integrates gestational-age-aware spatial constraints and morphological priors into a deep learning-based pose estimation pipeline. Evaluated on clinical 4D MRI datasets spanning early and late gestation, it significantly reduces pose estimation variance (p < 0.01) while improving accuracy and stability. This work overcomes the longstanding bottleneck in quantitative motion analysis of early-gestational 4D fetal imaging and provides a clinically deployable foundation for early intervention in fetal neurodevelopmental assessment.

Technology Category

• Computer Vision: Biometrics, Face, Gesture & Pose
• Intelligent Robots: State Estimation
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Web Mining and Content Analysis: Web data generation and simulation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Fetal motion is a critical indicator of neurological development and intrauterine health, yet its quantification remains challenging, particularly at earlier gestational ages (GA). Current methods track fetal motion by predicting the location of annotated landmarks on 3D echo planar imaging (EPI) time-series, primarily in third-trimester fetuses. The predicted landmarks enable simplification of the fetal body for downstream analysis. While these methods perform well within their training age distribution, they consistently fail to generalize to early GAs due to significant anatomical changes in both mother and fetus across gestation, as well as the difficulty of obtaining annotated early GA EPI data. In this work, we develop a cross-population data augmentation framework that enables pose estimation models to robustly generalize to younger GA clinical cohorts using only annotated images from older GA cohorts. Specifically, we introduce a fetal-specific augmentation strategy that simulates the distinct intrauterine environment and fetal positioning of early GAs. Our experiments find that cross-population augmentation yields reduced variability and significant improvements across both older GA and challenging early GA cases. By enabling more reliable pose estimation across gestation, our work potentially facilitates early clinical detection and intervention in challenging 4D fetal imaging settings. Code is available at https://github.com/sebodiaz/cross-population-pose.

Problem

• Estimating fetal pose across gestational ages robustly
• Overcoming anatomical changes in early gestation periods
• Generalizing pose models from late to early GA data

Innovation

• Cross-population data augmentation framework
• Fetal-specific augmentation simulating early GA
• Improves pose estimation across gestational ages