Shiraz University Fetal Heart Sounds Database

Abstract

The Shiraz University (SU) fetal heart sounds database (SUFHSDB) contains fetal and maternal phonocardiogram (PCG) recordings from 109 pregnant women in single and twin pregnancies. The recordings were made at Hafez Hospital of Shiraz University of Medical Sciences, Shiraz, Iran, from mothers aged from 16 to 47 years with a digital stethoscope placed on the lower maternal abdomen. In total, 99 subjects had one signal recorded, three subjects had two and seven cases of twins were recorded individually, resulting in 119 recordings. The average duration of each record is about 90 seconds. The sampling rate was generally 16,000 Hz with 16-bit quantization and a few recordings at 44,100 Hz. The data was recorded in wide-band mode of the digital stethoscope, with a frequency response of 20 Hz to 1 kHz.

Background

In high-risk pregnancies, regular monitoring with cardiotocography (CTG) is advised. Although CTG is known as a noninvasive method for fetal heart rate (FHR) monitoring, in risky pregnancies where continuous and long-term monitoring is needed, ultrasound waves can be harmful for the fetus. Nowadays, mobile health plays an important role in people’s lifestyle. Remote monitoring enables patients to record their biological signals at home and to submit this data to a medical specialist for remote diagnosis. This service may offer benefits in terms of convenience and cost. For remote monitoring to be successful, the devices used by patients would ideally be low-cost and simple to use. Fetal phonocardiography (PCG) is a promising technology in this field that may be used to monitor mechanical activity of the fetal heart. Due to its low-cost and passive technology (no energy is transmitted to the fetus during recording) it may be considered as an alternative for CTG for frequent and long-term measurement of the FHR. The objective of this study was to develop a fetal PCG database that could be used for the development and assessment of PCG-based algorithms in fetal cardiac monitoring studies [1,2].

Methods

The recordings were made at Hafez Hospital of Shiraz University of Medical Sciences, Shiraz, Iran, from 109 pregnant women (mothers aged from 16 to 47 years, mean±SD: 29.3±5.8 years with Body Mass Index (BMI) from 19.5 to 38.9 mean±SD: 29.2±4.0), with a digital JABES Electronic stethoscope (GS Technology Co. Ltd., South Korea) placed on the lower maternal abdomen as described in (Samieinasab and Sameni, 2015). In the case of twins (seven cases) the data were collected twice according to the locations advised by the expert gynecologist. Audacity® cross-platform audio software was used for recording and editing the signals on a Windows PC [3].

Data Description

In total, 99 subjects had one signal recorded, three subjects had two and seven cases of twins were recorded individually, resulting in 119 total recordings. The average duration of each record is about 90 seconds. The sampling rate was generally 16,000 Hz with 16-bit quantization and a few recordings at 44,100 Hz. The data was recorded in wide-band mode of the digital stethoscope, with a frequency response of 20 Hz to 1 kHz.

In most cases (91 subjects), the heart sounds of the mothers were also recorded before each fetal PCG recording session. As a result, a total number of 92 maternal heart sounds data are also available in the dataset (90 subjects with one heart sound signal and one subject with two signals recorded, resulting in 92 files in total).

The record names are in the form fX or mX where f stands for fetal recording, m stands for maternal recording, and X is the subject number. The data are in standard audio WAV (.wav) format and WFDB headers (.hea) are provided for reading the data description in WFDB format. An Excel spreadsheet (.xlsx) is also provided, which contains: subject ID, number of gravid/alive/abortions of the mother, maternal BMI, pregnancy term (weeks), gender of the fetus, mother’s age, clinical history (if available), average CTG heart-rate over 10-second windows (whenever available).

Usage Notes

Frequent and long-term monitoring of fetal health status is still a challenging task in high-risk pregnancies. The fetal PCG database was developed to help address this issue, offering a basis for the development and evaluation of algorithms for fetal cardiac monitoring. In Samieinasab et al, the data was used to develop a noninvasive method to extract fetal heart sound (FHS) from acoustic signals recorded from the maternal abdominal surface [1]. The proposed algorithm is based on single channel blind source separation (SCBSS), which utilizes empirical mode decomposition (EMD) and nonnegative matrix factorization (NMF) to extracts different sources from audio signal mixtures.

Release Notes

In this release, the average fetal heart rates acquired simultaneously via cardiotocography (CTG) over 10-second windows of some of the subjects is added to the spreadsheet, for reference.

Acknowledgements

The SUFHSDB was recorded as part of the Master's thesis of Ms. Maryam Samieinasab at Shiraz University, Shiraz, Iran and provided online by Dr. Reza Sameni during his faculty appointment at Shiraz University. The authors would like to thank Dr. M. Hosseiniasl and Ms. Nasihatkon from Shiraz Hafez Hospital for their valuable assistance during the acquisition of the fetal PCG records, and Dr. M.R. Samieinasab from Isfahan University of Medical Sciences, Isfahan, Iran, for his clinical advice and support throughout this study.

Conflicts of Interest

The authors declare to have no conflicts of interest.

References

1. M. Samieinasab and R. Sameni, Fetal phonocardiogram extraction using single channel blind source separation, 2015 23rd Iranian Conference on Electrical Engineering, Tehran, 2015, pp. 78-83. DOI: https://doi.org/10.1109/IranianCEE.2015.7146186
2. Chengyu Liu, David Springer, Qiao Li, Benjamin Moody, Ricardo Abad Juan, Francisco J Chorro, Francisco Castells, José Millet Roig, Ikaro Silva, Alistair E W Johnson, Zeeshan Syed, Samuel E Schmidt, Chrysa D Papadaniil, Leontios Hadjileontiadis, Hosein Naseri, Ali Moukadem, Alain Dieterlen, Christian Brandt, Hong Tang, Maryam Samieinasab, Mohammad Reza Samieinasab, Reza Sameni, Roger G Mark, and Gari D Clifford. An open access database for the evaluation of heart sound algorithms, Physiological Meassurement. 2016 Dec; 37(12):2181-2213. DOI: https://doi.org/10.1088/0967-3334/37/12/2181
3. Audacity, open source audio software. https://www.audacityteam.org/ [Accessed: 7 January 2021]