Abstract
Objective:
Technology in neurointensive care units can collect and store vast amounts of complex patient data. The CHART-ADAPT project is aimed at developing technology that will allow for the collection, analysis and use of these big data at the patient’s bedside in neurointensive care units. A requirement of this project is to automatically extract and transfer high-frequency waveform data (e.g. ICP) from monitoring equipment to high performance computing infrastructure for analysis. Currently, no agreed data standard exists in neurointensive care for the description of this type of data. In this pilot study, we investigated the use of Medical Waveform Format Encoding Rules (MFER—www.mfer.org-ISO 11073-92001) as a possible data standard for neurointensive care waveform data.
Materials and methods: Several waveform formats were explored (e.g. XML, DICOM waveform) and evaluated for suitability given existing computing infrastructure constraints, e.g. NHS network capacity and the processing capabilities of existing integration software. Key requirements of the format included a compact data size and the use of a recognised standard. The MFER waveform format (ISO/TS 11073-92001) met both requirements. To evaluate the practicality of the MFER waveform format, seven waveform signals (ICP, ECG, ART, CVP, EtCO2, Pleth, Resp) collected over a period of 8 h from a patient at the Institute of Neurological Sciences in Glasgow were converted into MFER waveform format.
Results: The MFER waveform format has two main components: sampling information and frame information. Sampling information describes the frequency of the data sampling and the resolution of the data. Frame information describes the data itself; it consists of three elements: data block (the actual data), channel (each type of waveform data occupies a channel) and sequence (the repetition of the data). All seven waveform signals were automatically and successfully converted into the MFER waveform format. One MFER file was created for each minute of data (total of 479 files, 181 KB each).
Conclusions: The MFER waveform format has potential as a lightweight standard for representing high-frequency neurointensive care waveform data. Further work will include a comparison with other waveform data formats and a live trial of using the MFER waveform format to stream patient data over a longer period.
References
Moss, L., Shaw, M., Piper, I., Hawthorne, C., Kinsella, J., Aridhia, Philips healthcare. Apache spark for the analysis of high frequency neurointensive care unit data: preliminary comparison of scala vs. R. Proceedings of American Medical Informatics Association 2016 Annual Symposium (AMIA 2016).
Moss, L., Shaw, M., Piper, I., Hawthorne, C., Kinsella, J., Aridhia, Philips healthcare. Enabling big data analysis in the neurointensive care unit. British Neurosurgical Research Group Meeting 2016, Cambridge, UK, March 2016.
Acknowledgements
CHART-ADAPT (http://www.chartadapt.org) is an Innovate UK co-funded project and the project partners are: Aridhia, Philips Healthcare, University of Glasgow and NHS Greater Glasgow and Clyde.
Conflicts of interest statement
We declare that we have no conflicts of interest.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG
About this paper
Cite this paper
Piper, I., Shaw, M., Hawthorne, C., Kinsella, J., Moss, L. (2018). Medical Waveform Format Encoding Rules Representation of Neurointensive Care Waveform Data. In: Heldt, T. (eds) Intracranial Pressure & Neuromonitoring XVI. Acta Neurochirurgica Supplement, vol 126. Springer, Cham. https://doi.org/10.1007/978-3-319-65798-1_38
Download citation
DOI: https://doi.org/10.1007/978-3-319-65798-1_38
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-65797-4
Online ISBN: 978-3-319-65798-1
eBook Packages: MedicineMedicine (R0)