A Dual Adversarial Calibration Framework for Automatic Fetal Brain Biometry
This work addresses the challenge of accurate fetal brain measurements in resource-limited settings, representing a domain-specific incremental improvement.
The paper tackles the problem of automatic fetal brain biometry for low-cost ultrasound images in low- and medium-income countries by proposing an unsupervised domain adaptation method, achieving mean absolute errors of 2.43mm and 1.65mm for transcerebellar diameter and head circumference, compared to 7.28mm and 5.65mm for state-of-the-art methods.
This paper presents a novel approach to automatic fetal brain biometry motivated by needs in low- and medium- income countries. Specifically, we leverage high-end (HE) ultrasound images to build a biometry solution for low-cost (LC) point-of-care ultrasound images. We propose a novel unsupervised domain adaptation approach to train deep models to be invariant to significant image distribution shift between the image types. Our proposed method, which employs a Dual Adversarial Calibration (DAC) framework, consists of adversarial pathways which enforce model invariance to; i) adversarial perturbations in the feature space derived from LC images, and ii) appearance domain discrepancy. Our Dual Adversarial Calibration method estimates transcerebellar diameter and head circumference on images from low-cost ultrasound devices with a mean absolute error (MAE) of 2.43mm and 1.65mm, compared with 7.28 mm and 5.65 mm respectively for SOTA.