More than dots on a map
Geographic information systems are used to display and monitor infectious disease data, often with additional layers of information relevant to the cause and nature of disease. We recently completed a decade-long analysis of an emerging infectious disease with colleagues in Sri Lanka. The molecular biology behind the story of emerging infection was handled with a GIS-compatible programme known as GENGIS. However, the plotting of patient locations was plotted manually to secure the evolving national picture over time, preventing a software driven density analysis of cases. We took a deep dive into the disease data using exploratory data analysis and visualisation tools before attempting multivariate analysis.
What we need for this and similar infectious disease monitoring is a map plotting tool that enables addition of new GIS-tagged data as it trickles through so that we can spot unusual patterns of spatial contouring such as increasing density (sometimes called a hot spot). A suite of optional features such as selection of an interesting cluster for histogram, tree classification, principal component analysis and vector-based methods of cluster analysis could be used.
We have performed our machine learning in Orange to date, and included the results of tree classification in this year's scientific report. Working in Orange has been a good starting point, and should present opportunities to assemble a machine learning workflow quickly and easily. Modification to existing applications to optimise their features will be easier than starting from scratch. We have CSV data sets and Excel spreadsheet data to work with.
References
-
Clinical, Bacteriologic, and Geographic Stratification of Melioidosis Emerges from the Sri Lankan National Surveillance Program. Sathkumara HD, Merritt AJ, Corea EM, Krishnananthasivam S, Natesan M, Inglis TJJ, De Silva AD. Am J Trop Med Hyg. 2018 Feb;98(2):607-615. doi: 10.4269/ajtmh.17-0441.
-
Field trials of blood culture identification FilmArray in regional Australian hospitals.
Bzdyl NM, Urosevic N, Payne B, Brockenshire R, McIntyre M, Leung MJ, Weaire-Buchanan G, Geelhoed E, Inglis TJJ. J Med Microbiol. 2018 Mar 13. doi: 10.1099/jmm.0.000714.
Client
Contact Person: Tim Inglis, Pathology & Laboratory Medicine, School of Medicine, UWA
Telephone: 0407 94 631
Email: [email protected]
Preferred method of contact: e-mail
Location: QEII Medical Centre, office in PathWest building, lab in L block Microbiology
Client Unavailability
None
IP Exploitation Model
The client wishes to use a Creative Commons CC BY-NC model to deal with IP embodied in the project.