PA²Net: Period-Aware Attention Network for Robust Fetal ECG Detection

The noninvasive fetal ECG (FECG) is used to monitor fetal well-being at prenatal and intrapartum. However, it is challenging to detect the FECG signal from the abdominal electrocardiogram (ECG) due to the following problems: 1) the FECG signal is weak and often masked by noise and 2) the FECG signal is mixed or overlapped with the maternal ECG (MECG) signal. To solve such problems, a period-aware attention network (PA²Net) is proposed for FECG detection by designing three modules, where an FECG period-aware attention module (FPAM), which is designed to suppress the noise interference by modeling the periods and features of signals, is first employed to detect FECG signals masked in noise; next, an MECG period-aware attention module, which is generated from the FPAM by the KL-divergence-based weight sharing module, can collaborate with the FPAM to search for mixed ECG signals of FECG and MECG. Finally, an antialiasing signal separation module is developed to estimate FECG signals from mixed ECG signals. Experiments conducted on the benchmarks show that the proposed PA²Net can achieve excellent performance with the PPV of 98.83%, 99.74%, 99.67%, and 98.47% on PCDB, ADFECGDB, NIFECGDB, and LHDB databases, respectively, allowing obstetricians to use PA²Net to monitor fetal health and diagnose diseases.

I. Introduction

Since 1970, electronic fetal monitoring (EFM) has been applied to monitor prenatal development [1]. However, the World Health Organization does not encourage to use EFM in parturition because invasive device infection may potentially occur during the monitoring process [2]. Therefore, it is essential to adopt noninvasive EFM tools to monitor fetal well-being at prenatal and intrapartum. As one of the effective noninvasive fetal monitoring methods, the fetal ECG (FECG) has attracted much attention because it can provide diagnostic information related to fetal physiological activities [3]. For example, the fetal heart rate (FHR) and fetal movements are associated with oxygen deprivation in utero [4]. Despite its powerful medical representation capabilities, the FECG signal always suffers from a lot of noise interferences, such as power-line interference, baseline drift, and the maternal ECG (MECG) signal, resulting in great confusion for obstetricians to monitor fetal health and diagnosing diseases.