Database Open Access
NInFEA: Non-Invasive Multimodal Foetal ECG-Doppler Dataset for Antenatal Cardiology Research
Danilo Pani , Eleonora Sulas , Monica Urru , Reza Sameni , Luigi Raffo , Roberto Tumbarello
Published: Nov. 12, 2020. Version: 1.0.0
When using this resource, please cite:
(show more options)
Pani, D., Sulas, E., Urru, M., Sameni, R., Raffo, L., & Tumbarello, R. (2020). NInFEA: Non-Invasive Multimodal Foetal ECG-Doppler Dataset for Antenatal Cardiology Research (version 1.0.0). PhysioNet. https://doi.org/10.13026/c4n5-3b04.
Sulas, E., Urru, M., Tumbarello, R., Raffo, L., Sameni, R., Pani, D., A non-invasive multimodal foetal ECG–Doppler dataset for antenatal cardiology research. Sci Data 8, 30 (2021). https://doi.org/10.1038/s41597-021-00811-3
Please include the standard citation for PhysioNet:
(show more options)
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.
The development of algorithms for the extraction of the foetal ECG (fECG) from non-invasive recordings is hampered by the lack of publicly-available reference datasets, which could be used to benchmark different algorithms while providing a ground truth on the foetal heart activity when an invasive scalp lead is unavailable. By enriching the electrophysiological recordings with simultaneous multimodal signals, these datasets could also help the investigation of the foetal cardiac physiology, providing ground truth for the analysis in early pregnancy, when the fECG is not directly accessible.
The Non-Invasive Multimodal Foetal ECG-Doppler Dataset for Antenatal Cardiology Research (NInFEA) is the first open-access dataset featuring simultaneous non-invasive electrophysiological recordings, fetal pulsed-wave Doppler (PWD) and maternal respiration signals. The dataset includes 60 entries from 39 voluntary pregnant women, between the 21st and the 27th week of gestation. Every entry is composed of 27 electrophysiological channels (2048 Hz, 22 bits, acquired by means of the TMSi Porti7 system), maternal respiration signal (through a resistive thoracic belt), synchronised foetal trans-abdominal PWD and clinical annotations provided by expert clinicians at the time of the signal collection.
To date, there are limited public datasets of non-invasive fECG acquired from the maternal abdomen, datasets with foetal scalp electrode reference during labour and Doppler ultrasound during pregnancy. However, the previous public datasets in this domain have some limitations for researchers working on simultaneous Doppler signals and signal processing techniques for non-invasive fECG extraction.
Prior datasets in this context have a limited number of channels or do not include any maternal reference; others present a limited number of signals, and to the best of our knowledge none of them provides ground truth about the foetal heart activity in early pregnancy. The Non-Invasive Multimodal Foetal ECG-Doppler Dataset for Antenatal Cardiology Research (NInFEA) aims at covering the shortcomings of currently available datasets and satisfying the aforementioned features, specifically for research purposes [1-4].
The dataset creation was approved by the Independent Ethical Committee of the Cagliari University Hospital (AOU Cagliari) and performed following the principles outlined in the 1975 Helsinki Declaration, as revised in 2000. All volunteers provided their signed informed consent to the protocol.
Before each recording session, the cardiologist performed the medical examination as per the current clinical guidelines and screened the morphology and the functionality of the foetal heart by foetal echocardiography. After the assessment of a healthy status of the foetus, the informed consent was signed by the volunteer pregnant woman and the acquisition setup was arranged.
Recordings were performed with the subject at rest, in a comfortable semi-sitting position on an echocardiography table. After the electrodes were attached, the respiration belt was fastened around the subject’s chest and the cardiologist checked the position of the foetus by B-mode echography (this information is included in the metadata of the dataset). Then, the simultaneous multimodal acquisition started, after the correct sample volume for the echocardiography was identified.
For the recording of the PWD signal, the five-chamber apical window was adopted. The Philips iE33 Ultrasound Machine (Philips Healthcare, The Netherlands) was used to perform the PWD measurement. The native resolution of the video was 1680 x 1050 pixels, at a frame rate of 60 Hz, and the sweep speed was set to 75 mm/s. The PWD signal was recorded through the DVI output with a USB3HDCAP USB3.0 Video Capture Device (StarTech, UK). This frame grabber was able to record 1080p HD videos with a frame-rate up to 60 frames per second and H.264 encoding. The whole video was converted into a single wide image by means of a Matlab custom tool.
For this dataset, the bio-potentials were recorded with the Porti7 portable physiological measurement system (TMSi, The Netherlands). It features simultaneous sampling up to 2048 Hz on the available input channels, but with an input bandwidth limited by the internal digital decimation filter to approximately 550 Hz. For the electrode placement topography, please refer to the corresponding publication on Scientific Data. Due to its high spatial redundancy, this configuration can be mapped to lower-dimensional schemes by spatial sub-sampling. A piezo-resistive respiration belt was placed around the maternal chest and was connected to one of the auxiliary inputs of the Porti7. The Porti7 device and the iE33 ultrasound machine cannot be directly synchronised for the acquisition of long traces. Synchronization was therefore performed by post-processing.
The dataset includes 60 entries from 39 voluntary pregnant women, between the 21st and the 27th week of gestation, living in Sardinia (Italy), and recorded at the Pediatric Cardiology and Congenital Heart Disease Unit, Brotzu Hospital, Cagliari, between 2017 and 2018. Signal length varies from 7.5 s to 119.8 s (average 30.6 s 20.6 s). For every data record, two files are available: the PWD trace as an image in the standard bitmap (.bmp) format and the electrophysiological and respiration signals in open binary format. The former was acquired by means of a five-chamber apical window with a Philips iE33 Ultrasound Machine (Philips Healthcare, The Netherlands). The latter was acquired with a Porti7 portable physiological measurement system (TMSi, The Netherlands) and exported in a binary format (.bin) described in the following table (IEEE little-endian, the 8-byte data words of all channels are written sample-wise starting from channel 1 to channel c).
|Description||# of bytes||precision|
|Sampling frequency||8||floating-point double precision|
|# of channels (r)||8||unsigned integer 64-bit|
|# of time samples (c)||8||unsigned integer 64-bit|
|Data||floating-point double precision|
In these files, the first 24 rows are associated to unipolar channels, gathering the signals from 24 electrodes placed on the maternal abdomen and back, rows from 25 to 27 are associated to differential channels gathering the signals from six electrodes positioned on the maternal thorax, for recording the maternal ECG, rows from 28-31 are associated to unused channels, row 32 is the maternal respiration signal, row 33 is an internal saw-tooth signal and row 34 is associated to the trigger signal used for synchronisation.
A small library of Matlab (The Mathworks, MA, USA) custom functions accompanies the dataset. In particular, a binary file reader for Matlab is provided, enabling to load the signals acquired with the Porti7 electrophysiological recording system in a Matlab variable. Moreover, a graphical user interface enables simultaneously scrolling of the long PWD image (first loaded as .bmp file) and all the related Porti7 channels and internal signal (in .bin format).
Interested users can modify its source code in order to show the fECG signals extracted by the preferred method without any limitation. Finally, Matlab code to extract the PWD envelope is also available for PWD processing. It provides the upper and lower envelope from the long .bmp PWD images, as described in the referenced papers on this topic [5-7].
The dataset and processing tools are provided for scientific research purposes and may be used with proper citation to the indicated sources. Further processing tools can be obtained from .
All the signals come from cardiologically-healthy foetuses.
The authors wish to thank the Pediatric Cardiology and Congenital Heart Disease Unit, Brotzu Hospital (Cagliari, Italy), where the dataset was collected, and all the voluntary pregnant women for their kindness in giving their signals for this research. The authors gratefully thank Alessandra Cadoni, Graziella Secchi, Luisa Aru, Elisa Farris, Chiara Fenu, Elisa Gusai, Giulia Baldazzi, Giulia Pili for their support in the recording of the signals included in this dataset.
Part of this research was supported by the Italian Government—Progetti di InteresseNazionale (PRIN) under the grant agreement 2017RR5EW3 - ICT4MOMs project.
Eleonora Sulas is grateful to Sardinia Regional Government for supporting her PhD scholarship (P.O.R.F.S.E., European Social Fund 2014-2020).
Reza Sameni acknowledges the funding from the European Research Council Advanced Grant Number 320684, on Challenges in the Extraction and Separation of Sources (CHESS) for his contribution in this research, provided during his appointment at GIPSA-lab, Grenoble Alpes University, Grenoble, France.
Conflicts of Interest
The authors declare the absence of conflicts of interests.
- Sulas E, Ortu E, Raffo L, Urru M, Tumbarello R, Pani D, Automatic Recognition of Complete Atrioventricular Activity in Fetal Pulsed-Wave Doppler Signals, 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Honolulu, HI, 2018, pp. 917-920, doi: 10.1109/EMBC.2018.8512329.
- Sulas E, Ortu E, Urru M, Cadoni A, Tumbarello R, Raffo L, Pani D, Fetal Pulsed-Wave Doppler Atrioventricular Activity Detection by Envelope Extraction and Processing, 2018 Computing in Cardiology Conference (CinC), Maastricht, Netherlands, 2018, pp. 1-4, doi: 10.22489/CinC.2018.361
- Sulas E, Urru M, Tumbarello R, Raffo L, Pani D, Automatic detection of complete and measurable cardiac cycles in antenatal pulsed-wave Doppler signals, Computer Methods and Programs in Biomedicine, Volume 190, 2020, 105336, ISSN 0169-2607, https://doi.org/10.1016/j.cmpb.2020.105336.
- Sulas E, Urru M, Tumbarello R, Raffo L, Pani D, Comparison of Single- and Multi-reference QRD-RLS adaptive filter for non-invasive fetal electrocardiography, 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Berlin, Germany, 2019, pp. 1-5, doi: 10.1109/EMBC.2019.8856824.
- Sulas, E., Urru, M., Tumbarello, R., Raffo, L., & Pani, D. (2019). Systematic analysis of single- and multi-reference adaptive filters for non-invasive fetal electrocardiography. Mathematical biosciences and engineering : MBE, 17(1), 286–308. https://doi.org/10.3934/mbe.2020016
- Baldazzi G, Sulas E, Brungiu E, Urru M, Tumbarello R, Raffo L, Pani D, Wavelet-Based Post-Processing Methods for the Enhancement of Non-Invasive Fetal ECG, 2019 Computing in Cardiology (CinC), Singapore, Singapore, 2019, pp. 1-4, doi: 10.23919/CinC49843.2019.9005921.
- Baldazzi G, Sulas E, Urru M, Tumbarello R, Raffo L, Pani D, Wavelet denoising as a post-processing enhancement method for non-invasive foetal electrocardiography, Computer Methods and Programs in Biomedicine, Volume 195, 2020, 105558, ISSN 0169-2607, https://doi.org/10.1016/j.cmpb.2020.105558.
- Sameni R, The Open-Source Electrophysiological Toolbox (OSET), version 3.14, 2018. [Online]. Available: http://www.oset.ir (https://gitlab.com/rsameni/OSET)
Anyone can access the files, as long as they conform to the terms of the specified license.
License (for files):
Open Data Commons Attribution License v1.0
Total uncompressed size: 2.0 GB.
Access the files
- Download the ZIP file (792.9 MB)
Download the files using your terminal:
wget -r -N -c -np https://physionet.org/files/ninfea/1.0.0/
|Interfaccia_dataset.m (download)||16.1 KB||2020-11-12|
|Logo_UniCa.jpg (download)||377.7 KB||2020-10-05|
|ReadBinaryFile.m (download)||1.3 KB||2020-11-11|
|envelope_extraction.m (download)||3.3 KB||2020-11-12|