Database Open Access

# Indian Institute of Science Fetal Heart Sound Database (IIScFHSDB)

Published: Sept. 15, 2022. Version: 1.0

Bhaskaran, A., & Arora, M. (2022). Indian Institute of Science Fetal Heart Sound Database (IIScFHSDB) (version 1.0). PhysioNet. https://doi.org/10.13026/9vvw-cx05.

Amrutha, B; Sidhesh Kumar, J; George, S. & Arora, M. Heart rate estimation and validation algorithm for fetal phonocardiography Physiological Measurement, 2022

Goldberger, A., Amaral, L., Glass, L., Hausdorff, J., Ivanov, P. C., Mark, R., ... & Stanley, H. E. (2000). PhysioBank, PhysioToolkit, and PhysioNet: Components of a new research resource for complex physiologic signals. Circulation [Online]. 101 (23), pp. e215–e220.

## Abstract

The Indian Institute of Science Fetal Heart Sound Database (IIScFHSDB) has 60 fetal phonocardiography (fPCG) recordings obtained from expecting mothers of age 18-27 years of age with gestation weeks 30-40. The recordings are made at St. John’s Hospital, Bangalore with an electronic stethoscope. The data is collected from the antenatal and labor wards by placing the electronic stethoscope on the lower abdomen of the subject. The fPCG recordings are obtained from 60 subjects with the gain of 500X and the average duration of the recordings is 8 minutes with a sampling frequency of 2 kHz.

## Background

Fetal monitoring during pregnancy and labor helps in the identification of fetuses at high risk. The most monitored fetal parameter is the Fetal Heart Rate (FHR), and one can obtain clinically relevant information such as baseline FHR, acceleration, and deceleration from an FHR trace. There are various techniques to determine the FHR, namely fetal doppler ultrasound, direct and indirect fetal electrocardiography (fECG), fetal magnetocardiography (fMCG), and fetal phonocardiography (fPCG). The main advantage of the devices based on fPCG are non-invasive [1], passive, simple, and easy to use . One of the main challenges faced in FHR extraction from fPCG is the difficulty in extracting information from the noisy data. If we can obtain reliable FHR measurements from fPCG, then the devices based on fPCG will have widespread use in resource-constrained settings. Please note that in certain resource-constrained settings, the amount of external noise is also much higher due to crowding and the lack of private wards. The main objective of this database is to provide fPCG recordings obtained from a noisy clinical setting that can be used for the development and evaluation of various signal processing algorithms for fPCG denoising and FHR determination [2].

## Methods

The fPCG recordings are obtained using an electronic stethoscope (SS30LA) connected to a data acquisition system (MP36, Biopac Systems Inc.) with a gain of 500X and a sampling frequency 2 kHz. The gain of 500X implies the input signal is amplified 500 times before digitization, i.e. ratio of the output signal to the input signal is 500 (dimensionless). The data is recorded with Biopac Student Lab software and then exported as an audio file (.wav) which can be read in MATLAB for further analysis. The data is obtained from pregnant women with gestation weeks greater than 30 and singleton pregnancies, with no known maternal complications. The data is collected from the location advised by a trained doctor. The demographic and clinical information of the 60 subjects is shown in the table below:

 Age Gestation week Weight (kg) Height (cm) Mean ± SD 25±3.7 37±2.0 68.3±10.9 155±7.3 Range 18-37 31-40 49-103 144-171

## Data Description

In total, the 60 fPCG recordings are collected from 60 subjects. The average recording time is around 8 minutes. The data files are in wave (.wav) format that can be easily analyzed using MATLAB.No prior filtering of the data is done except for the inbuilt notch filter in the biopac to remove the line frequency.

The record is labeled as subject_x. The average recording time of the data is 8 minutes. A .csv file is provided which gives information on the subjects namely height, weight, gravida, age, and clinical conditions if present along with FHR values noted from the patient file when available.

## Usage Notes

The fPCG data was developed to encourage the development and evaluation of algorithms for monitoring cardiac activities of the fetus, particularly in the case when noise plays a role during data acquisition. In [2], the data was used to develop the FHR extraction algorithm which uses autocorrelation and FHR validation steps to accurately measure the FHR for different signal qualities. example_code.m provides the procedures for reading the audio file and applying comb filter using MATLAB.

## Release Notes

Version 1.0: Initial Release

## Ethics

The study was approved by the Institutional Ethics Committee,St. John’s Hospital, Bangalore (IEC code No. 07/2018). The data is collected after obtaining informed consent.

## Acknowledgements

We would like to thank Dr. Shirley George and doctors at the Department of Obstetrics and Gynaecology, St. John’s hospital, Bangalore for support in data acquisition.

## Conflicts of Interest

The authors declare no conflict of interest.

## References

1. Ibrahim, Emad A., et al. "A comparative study on fetal heart rates estimated from fetal phonography and cardiotocography." Frontiers in physiology 8 (2017): 764.
2. Amrutha B et.al ,”Heart Rate Estimation and Validation Algorithm for Fetal Phonocardiography”, physiological measurements.

##### Access

Access Policy:
Anyone can access the files, as long as they conform to the terms of the specified license.

##### Corresponding Author
You must be logged in to view the contact information.

## Files

Total uncompressed size: 377.2 MB.

##### Access the files
wget -r -N -c -np https://physionet.org/files/fetalheartsounddata/1.0/

Name Size Modified