Adaptive 3D Localization of 2D Freehand Ultrasound Brain Images

Abstract

freehand ultrasound is the mainstay in prenatal care and fetal growth monitoring. The task of matching corresponding cross-sectional planes in the 3D anatomy for a given 2D ultrasound brain scan is essential in freehand scanning, but challenging. We propose AdLocUI, a framework that Adaptively Localizes 2D Ultrasound Images in the 3D anatomical atlas without using any external tracking sensor. We first train a convolutional neural network with 2D slices sampled from co-aligned 3D ultrasound volumes to predict their locations in the 3D anatomical atlas. Next, we fine-tune it with 2D freehand ultrasound images using a novel unsupervised cycle consistency, which utilizes the fact that the overall displacement of a sequence of images in the 3D anatomical atlas is equal to the displacement from the first image to the last in that sequence. We demonstrate that AdLocUI can adapt to three different ultrasound datasets, acquired with different machines and protocols, and achieves significantly better localization accuracy than the baselines. AdLocUI can be used for sensorless 2D freehand ultrasound guidance by the bedside. The source code is available at https://github.com/pakheiyeung/AdLocUI.

Ultrasound Plane Pose Regression: Assessing Generalized Pose Coordinates in the Fetal Brain

“…Despite our promising results relative to the existing literature [8]- [10], our regression network still produces outlier predictions with large errors. This is to be expected, given that we perform estimations based on single US scans, which can be noisy.…”

“…The model was tested on fetal brain volumes with a GA ranging from 18 to 22 weeks. Then, in [10], an unsupervised cycle consistency using the fact that the overall displacement of a sequence of images in the 3D anatomical atlas is equal to the displacement from the first image to the last in that sequence was added.…”

“…We evaluate to which extent the brains of different fetuses can be registered to the same generalized coordinate frame. While this is essential to assess training and validation data for fetal brain pose regression models, it has not been done in the previous literature on this topic[8]-[10].…”