Temporal Convolution Networks for Real-Time Abdominal Fetal Aorta Analysis with Ultrasound

TL;DR

This paper introduces a neural network architecture combining convolutional layers, C-GRU, and a cyclic loss function to accurately and efficiently measure fetal abdominal aorta diameter from ultrasound videos in real-time.

Contribution

The proposed architecture significantly improves measurement accuracy and speed, enabling real-time analysis of ultrasound sequences for fetal health assessment.

Findings

Mean squared error reduced from 0.31 mm^2 to 0.09 mm^2

Relative error decreased from 8.1% to 5.3%

Achieves 289 frames per second for real-time use

Abstract

The automatic analysis of ultrasound sequences can substantially improve the efficiency of clinical diagnosis. In this work we present our attempt to automate the challenging task of measuring the vascular diameter of the fetal abdominal aorta from ultrasound images. We propose a neural network architecture consisting of three blocks: a convolutional layer for the extraction of imaging features, a Convolution Gated Recurrent Unit (C-GRU) for enforcing the temporal coherence across video frames and exploiting the temporal redundancy of a signal, and a regularized loss function, called \textit{CyclicLoss}, to impose our prior knowledge about the periodicity of the observed signal. We present experimental evidence suggesting that the proposed architecture can reach an accuracy substantially superior to previously proposed methods, providing an average reduction of the mean squared error from $0.31 mm^2$ (state-of-art) to $0.09 mm^2$, and a relative error reduction from $8.1\%$ to $5.3\%$. The mean execution speed of the proposed approach of 289 frames per second makes it suitable for real time clinical use.