gms | German Medical Science

62. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e. V. (GMDS)

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie

17.09. - 21.09.2017, Oldenburg

Visualizing patient cohort data in tranSMART: a phenotype toolbox

Meeting Abstract

Search Medline for

  • Robert Lodahl - Universitätsmedizin Göttingen, Göttingen, Deutschland
  • Christian R Bauer - Universitätsmedizin Göttingen, Göttingen, Deutschland
  • Benjamin Baum - Universitätsmedizin Göttingen, Göttingen, Deutschland
  • Ulrich Sax - Universitätsmedizin Göttingen, Göttingen, Deutschland

Deutsche Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie. 62. Jahrestagung der Deutschen Gesellschaft für Medizinische Informatik, Biometrie und Epidemiologie e.V. (GMDS). Oldenburg, 17.-21.09.2017. Düsseldorf: German Medical Science GMS Publishing House; 2017. DocAbstr. 256

doi: 10.3205/17gmds151, urn:nbn:de:0183-17gmds1512

Published: August 29, 2017

© 2017 Lodahl et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at http://creativecommons.org/licenses/by/4.0/.


Outline

Text

Introduction: Research platforms are one solution in current medical research to make use of large and diverse data sets [1]. Some platforms offer visualization tools to easily assess and access the data. Mostly the visualization efforts are focused on omics data, while phenotype visualization remains mostly basic [2]. To enhance the support for phenotype visualization, the goal of this work is to create a toolbox [3] offering helpful advanced visualization for phenotype data. As a starting point we included a heatmap visualization in tranSMART aiming on phenotype and other non-omics data.

Methods: The research platform tranSMART is used as data provider. Its data analysis plugin SmartR provides user interface and Rserve access [4] and serves as the software environment basis for our heatmap plugin. To avoid unnecessary development the genotype heatmap of SmartR is used as a template. This setup was also chosen to establish an equal user experience to other toolboxes [3] and create mutual benefit with other plugins for SmartR.

The development was carried out in two phases. In the first phase the statistical calculations were created, which are executed on the platform server using the statistical language R. In the second phase the user interface was adapted to allow the input of categorical and numeric phenotype data using the JavaScript-based SmartR frontend. Operational iterations of the plugin were tested by using real world data.

The data was based on a dataset generated according to specifications in §21 KHEntgG [5]. It contains clearance-relevant patient data including basic claims data and diagnoses coded in ICD-10-GM.

Results: A new SmartR plugin for tranSMART was created. This plugin allows it to correlate two sets of phenotype data with two options for heatmap coloring: (1) a numerical parameter of the patient dataset and (2) the absolute or relative number of patients in the dataset. Numerical concepts can easier be visualized with the optional binning function. The user decides which statistical method is used for the creation of the correlation. The correlation calculated using this configuration is displayed to the user in form of a heatmap. To enhance the usability of the plugin functions from SmartRs own genotype heatmap plugin are still available, e.g. sorting and ad hoc clustering resulting in tree diagrams.

Discussion: While we created a functional plugin for tranSMART with our initial goals reached, development is still ongoing. Extension of the plugin with additional functionality and improvements in user workflow are planned. One intended feature is an interactive “zoom” functionality.

For example: a heatmap is created correlating the age of patients and their ICD-10-GM diagnose in chapters (I to XXII). By selecting one distinct chapter another heatmap is created, showing the distribution of patients in every subcategory of the chapter (e.g. chapter I with blocks A00-B99).

The aim of the phenotype toolbox is not limited to these extensions of the phenotype heatmap, but is going to incorporate additional plugins for the visualization and access of phenotype data. These result are interesting for questions of feasibility, perhaps recruitment support and for medical controlling purposes.

Acknowledgements: This work was supported by the German Federal Ministry of Education and Research (BMBF) within the framework of the research and funding concepts e:Med (01ZX1306C/sysINFLAME) and i:DSem (031L0024A/MyPathSem).

Die Autoren geben an, dass kein Interessenkonflikt besteht.

Die Autoren geben an, dass kein Ethikvotum erforderlich ist.


References

1.
Kohane IS, Churchill SE, Murphy SN. A translational engine at the national side: informatics for integrating biology and the bedside. J AM Med Inform Assoc. 2012;19(2):181-5. DOI: 10.1136/amiajnl-2011-000492 External link
2.
Dunn W Jr, et al. Exploring and visualizing multidimensional data in translational research platforms. Brief Bioinform. 2016:bbw080. DOI: 10.1093/bib/bbw080 External link
3.
Bauer CR, et al. Interdisciplinary approach towards a systems medicine toolbox using the example of inflammatory diseases. Brief Bioinform. 2016:bbw024. DOI: 10.1093/bib/bbw024 External link
4.
Herzinger S, et al. SmartR: An open-source platform for interactive visual analytics for translational research data. Bioinformatics. 2017. DOI: 10.1093/bioinformatics/btx137 External link
5.
Deutscher Bundestag. Gesetz über die Entgelte für voll- und teilstationäre Krankenhausleistungen (Krankenhausentgeltgesetz - KHEntgG); Krankenhausentgeltgesetz vom 23. April 2002 (BGBl. I S. 1412, 1422), das durch Artikel 4 des Gesetzes vom 10. Dezember 2015 (BGBl. I S. 2229) geändert worden ist. URL: http://www.gesetze-im-internet.de/khentgg/index.html External link