MobileUNet-FPN: A Semantic Segmentation Model for Fetal Ultrasound Four-Chamber Segmentation in Edge Computing Environments

The apical four-chamber (A4C) view in fetal echocardiography is a prenatal examination widely used for the early diagnosis of congenital heart disease (CHD). Accurate segmentation of A4C key anatomical structures is the basis for automatic measurement of growth parameters and necessary disease diagnosis. However, due to the ultrasound imaging arising from artefacts and scattered noise, the variability of anatomical structures in different gestational weeks, and the discontinuity of anatomical structure boundaries, accurately segmenting the fetal heart organ in the A4C view is a very challenging task. To this end, we propose to combine an explicit Feature Pyramid Network (FPN), MobileNet and UNet, i.e., MobileUNet-FPN, for the segmentation of 13 key heart structures. To our knowledge, this is the first AI-based method that can segment so many anatomical structures in fetal A4C view. We split the MobileNet backbone network into four stages and use the features of these four phases as the encoder and the upsampling operation as the decoder. We build an explicit FPN network to enhance multi-scale semantic information and ultimately generate segmentation masks of key anatomical structures. In addition, we design a multi-level edge computing system and deploy the distributed edge nodes in different hospitals and city servers, respectively. Then, we train the MobileUNet-FPN model in parallel at each edge node to effectively reduce the network communication overhead. Extensive experiments are conducted and the results show the superior performance of the proposed model on the fetal A4C and femoral-length images.

I. Introduction

Congenital Heart Disease (CHD) is a type of heart condition that can be detected as early as the first trimester of life [1]. The defect is characterized by abnormalities in the heart structure, which can be mild or severe. CHD is the leading cause of infant mortality, estimated to cause approximately 4–13 deaths per 1000 live births [2]–[4]. Screening and diagnosing CHD during pregnancy is important so that life-saving treatment can be carried out in time. Newborns with severe heart disease who are not examined and diagnosed before birth may face worsening disease symptoms in the delivery room or in the first month of life. Echocardiography is the primary examination for the diagnosis of CHD. It is non-invasive, low-cost, convenient, and suitable for real-time medical imaging. The basic fetal heart screening is interpreted by the apical four-chamber (A4C) view of fetal echocardiography. In fact, although some fetal diseases can be directly visualized in the A4C view, some functional CHD cases cannot be observed, and require organ segmentation and parametric measurements to diagnose. The A4C view of a fetal ultrasound image and the masks of the key anatomical structures are shown in Fig. 1.