About the approach of solving machine learning problems integrated with data from open source systems of electronic medical records https://doi.org/10.33108/visnyk_tntu2019.03.105
Article Sidebar
Main Article Content
Abstract
In recent decades, open source health solutions and commercial tools have been actively
developed. The most common open source electronic health accounting systems are WorldVistA, OpenEMR and
OpenMRS. Scientists drew attention to the prospects of open-source electronic health records software and free
systems for countries with certain financial difficulties and such developing countries. Setting the task of machine
learning in medical research has been carried out. The flowchart presented in the paper demonstrates the main
steps for developing a machine learning model. It is noted that the task of importing training, testing and
forecasting data sets from EMR systems in the machine learning environment is not so trivial for a number of
reasons discussed in the study. Here are some basic approaches for accessing patient medical record data in
conventional EMR systems. Some features of approaches for the two most common EMR open source systems are
presented: OpenEMR, OpenMRS. Despite a long period of development and applications, even leading and
widespread EMR systems (both commercial and free open source) have limited or partial support for HL7
capabilities. Despite the challenges that the implementation level is considering, there are enough arguments to
adapt the use of data formats compatible with HL7 and to develop information systems that are machine learning
oriented. Experimental studies are related to the prediction of fractures for middle-aged women, confirm that this
is a pressing, preventive problem today. The development of the machine learning model is implemented in the
free software environment R, using the mlr package. As a result, we get machine learning models based on five
methods. The results of the effectiveness of the methods, using the mmce measure, show that the exact model of
compliance with the assessment of prediction quality is the random forest method, worst of all is the ferms method.
Article Details
Issue
Section
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
1. List of open-source health software. Electronic health or medical record. URL: https://en.wikipedia.org/wiki/List_of_open-_source_health_softwareElectronic_health_or_medical_record (accessed: 2017.11.12).
2. Almeida J., Frade S., Cruz-Correia R. Exporting Data from an openEHR Repository to Standard Formats. Conference on ENTERprise Information Systems / ProjMAN 2014 – International Conference on Project MANagement / HCIST 2014 International Conference on Health and Social Care Information Systems and Technologies. 2014. URL: https://www.sciencedirect.com/science/article/pii/S2212017314003843.
3. Aminpour F., Ahamdi M. Utilization of open source electronic health record around the world: A systematic review. Journal of research in medical sciences: the official journal of Isfahan University of Medical Sciences. 2014. № 19 (1). Р. 57.
4. Fritz F., Tilahun B., Dugas M. Success criteria for electronic medical record im- plementations in low- resource settings: a systematic review. Journal of the American Medical Informatics Association. 2015. № 22 (2). Р. 479–488. https://doi.org/10.1093/jamia/ocu038
5. Martsenyuk V., Semenets A. On code refactoring for decision making component combined with the open- source medical information system. Advances in Soft and Hard Computing AISC 889. 2019. URL: https://doi.org/10.1007/978-3-030-03314-9.
6. Martsenyuk V., Vakulenko D., Vakulenko L., Kos-Witkowska A. Information system of arterial oscillography for primary diagnostics of cardiovascu- lar diseases. Computer Information Systems and Industrial Management. CISIM 2018. Lecture Notes in Computer Science. 2018. Р. 46–56. URL: https://doi.org/10.1007/978-3-319-99954-85. https://doi.org/10.1007/978-3-319-99954-8_5
7. Reynolds C., Wyatt J. Open source, open standards, and health care information systems. Journal of medical Internet research. 2011. № 13 (1). https://doi.org/10.2196/jmir.1521
8. Semenets A. On organizational and methodological approaches of the emr-systems implementation in public health of Ukraine. Medical Informatics and Engineering. 2013. № 3.
9. Semenets A. About experience of the patient data migration during the open source emr-system implementation. Medical Informatics and Engineering. 2015. № 1.