Automatic Quality Assessment of First Trimester Crown-Rump-Length Ultrasound Images

TL;DR

This paper introduces a deep learning approach combining CNN and ViT for segmenting fetal structures in ultrasound images to verify image quality and adherence to clinical guidelines for accurate gestational age estimation.

Contribution

A novel CNN/ViT-based segmentation and mapping method that outperforms UNet and classification CNNs in assessing ultrasound image quality for fetal CRL measurement.

Findings

Proposed method outperforms UNet in segmentation accuracy.

Mapping approach is more accurate and explainable than classification CNNs.

Effective verification of ultrasound image quality for fetal age estimation.

Abstract

Fetal gestational age (GA) is vital clinical information that is estimated during pregnancy in order to assess fetal growth. This is usually performed by measuring the crown-rump-length (CRL) on an ultrasound image in the Dating scan which is then correlated with fetal age and growth trajectory. A major issue when performing the CRL measurement is ensuring that the image is acquired at the correct view, otherwise it could be misleading. Although clinical guidelines specify the criteria for the correct CRL view, sonographers may not regularly adhere to such rules. In this paper, we propose a new deep learning-based solution that is able to verify the adherence of a CRL image to clinical guidelines in order to assess image quality and facilitate accurate estimation of GA. We first segment out important fetal structures then use the localized structures to perform a clinically-guided mapping that verifies the adherence of criteria. The segmentation method combines the benefits of Convolutional Neural Network (CNN) and the Vision Transformer (ViT) to segment fetal structures in ultrasound images and localize important fetal landmarks. For segmentation purposes, we compare our proposed work with UNet and show that our CNN/ViT-based method outperforms an optimized version of UNet. Furthermore, we compare the output of the mapping with classification CNNs when assessing the clinical criteria and the overall acceptability of CRL images. We show that the proposed mapping is not only explainable but also more accurate than the best performing classification CNNs.