Dr Earl Duncan

Qualifications

PhD Math (Statistics), QUT.
B AppSc (Hons) (Mathematics), QUT.
B Math, QUT.
B Bus (Accounting, International Business), QUT.

Research Interests

• Spatial and spatio-temporal modelling (e.g. disease mapping)
• Data visualisation (e.g. choropleth maps, cartograms, VR and AR, interactive visualisations)
• Bayesian estimation techniques (improvements to existing algorithms, ensemble algorithms)
• Applications in:
  • Public health and epidemiology
  • Environmental sustainability
  • Hemeroby (human influence on the natural environment)
  • Urban planning
  • Geographic information systems (GIS)

Recent Projects

In 2022, I started work with the Australian Department of Health (Canberra) to develop and improve a supply and demand model for GPs, and subsequently other health force specialists.

In 2021, I worked at the Queensland cancer registry (Cancer Alliance) to help automate the tedious task of coding cancers, reporting, and data quality assurance using SQL and machine learning methods.

In early 2021, I worked with the Vic Department of Health to explore the spatial patterns and epidemiological relationships between COVID-19 vaccinations and vaccination sites across Victoria to better understand the effectiveness of existing vaccination sites, determine the factors driving vaccine uptake, and providing tools to assist decision makers in planning new vaccination sites. A key component of the analyses was the use of isochrones to realistically measure the burden of travel to a vaccination site.

In 2018 through 2020, I helped develop the Australian Cancer Atlas.  The output is a publicly accessible, online, interactive map (think Google Earth) showing estimated incidence and relative survival rates for many different types of cancers across Australia at the small area level (SA2s).  This was joint work with the Cancer Council Queensland and the QUT Visualisation and eResearch (ViseR) team to combine our expertise in statistical spatial modelling, data visualisation, and UX and UI design. A screen-capture of the atlas is shown below.

Publications

PhD Thesis:

Duncan, E. W.  2017.  Bayesian approaches to issues arising in spatial modelling.  PhD Thesis, Queensland University of Technology.  URL: https://eprints.qut.edu.au/112356/1/Earl_Duncan_Thesis.pdf.

Other Publications (most recent first):

1. Jahan, F., D. W. Kennedy, E. W. Duncan, and K. L. Mengersen.  2022.  Evaluation of spatial Bayesian Empirical Likelihood models in analysis of small area data.  PLOS ONE 17 (5): e0268130.  DOI: 10.1371/journal.pone.0268130.
2. Duncan, E.  2022.  Report on the Feasibility and Success of Auto-Coding Cancers in the QCCAT Virtual Cancer Registry: technical report.  Brisbane: Cancer Alliance, Queensland Department of Health. (Not publicly available).
3. Duncan, E. W. and J. Canevari.  2021.  An Investigation of COVID-19 Vaccination Service Coverage and Factors Influencing Vaccine Uptake, 30 June 2021: technical report.  Victoria: Analytics, Epidemiology and Analysis, Department of Health and Human Services (DHHS). (Not publicly available).
4. Cramb, S. M., E. W. Duncan, J. F. Aitken, H. P. Soyer, K. L. Mengersen, and P. D. Baade.  2020.  Geographical patterns in melanoma incidence across Australia: can thickness differentials explain the key drivers? Annals of Cancer Epidemiology 4.  DOI: 10.21037/ace-20-13.
5. Jahan, F., E. W. Duncan, S. M. Cramb, P. D. Baade, and K. L. Mengersen. 2020. Multivariate Bayesian meta-analysis: joint modelling of multiple cancer types using summary statistics.  International Journal of Health Geographics 19 (42).  DOI: 0.1186/s12942-020-00234-0.
6. Aswi, A., S. Cramb, E. Duncan, and K. Mengersen.  2020.  Evaluating the impact of a small number of areas on spatial estimation. International Journal of Health Geographics 19 (39).  DOI: 10.1186/s12942-020-00233-1.
7. Duncan, E. W., S. M. Cramb, P. D. Baade, K. L. Mengersen, T. Saunders, and J. F. Aitken.  2020.  A Practical Guide to Developing a Cancer Atlas using Bayesian Spatial Modelling.  Brisbane: Queensland University of Technology (QUT) and Cancer Council Queensland.  eBook available online: https://atlas.cancer.org.au/developing-a-cancer-atlas.
8. Jahan, F., E. W. Duncan, S. M. Cramb, P. D. Baade, and K. L. Mengersen.  2020.  Augmenting disease maps: a Bayesian meta-analysis approach.  Royal Society Open Science 7 (8): 192151.  DOI: 10.1098/rsos.192151.
9. Cramb, S., E. Duncan, P. Baade, and K. L. Mengersen. 2020. “A comparison of Bayesian spatial models for cancer incidence at a small area level: theory and performance”.  In Case Studies in Applied Bayesian Data Science, Volume 2259 of the Lecture Notes in Mathematics series, edited by K. L. Mengersen, P. Pudlo, and C. P. Robert, pp. 245-274. DOI: 10.1007/978-3-030-42553-1.
10. Duncan, E. W. and K. L. Mengersen.  2020. Comparing Bayesian spatial models: goodness-of-smoothing criteria for assessing under- and over-smoothing. PLOS ONE 15 (5): e0233019.  DOI: 10.1371/journal.pone.0233019.
11. Aswi, A., S. Cramb, E. Duncan, W. Hu, G. White, and K. Mengersen.  2020.  Climate variability and dengue fever in Makassar, Indonesia: Bayesian spatio-temporal modelling.  Spatial and Spatio-temporal Epidemiology 33: 100335.  DOI: 10.1016/j.sste.2020.100335.
12. Aswi, A., S. Cramb, E. Duncan, W. Hu, G. White, and K. Mengersen. 2020. Bayesian spatial survival models for hospitalisation of dengue: A case study of Wahidin hospital in Makassar Indonesia. International Journal of Environmental Research and Public Health 17 (3).  DOI: 10.3390/ijerph17030878.
13. Duncan, E. W., S. M. Cramb, J. F. Aitken, K. L. Mengersen, and P. D. Baade. 2019. Development of the Australian Cancer Atlas: Spatial Modelling, Visualisation, and Reporting of Estimates.  International Journal of Health Geographics 18 (21).  DOI: 10.1186/s12942-019-0185-9.
14. Mengersen, K., E. Duncan, J. Arbel, C. Alston-Knox, and N. White.  2019.  “Chapter 15: Applications in Industry”.  In Handbook of Mixture Analysis, Handbooks of Modern Statistical Methods Series, edited by S. Früwirth-Schnatter, G. Celeux, and C. P. Robert.  Milton: Chapman & Hall/CRC.
15. Cramb, S. M., E. W. Duncan, K. L. Mengersen, and P. D. Baade.  2018.  Australian Cancer Atlas, small-area incidence: technical report.  Brisbane: Cancer Council Queensland and Queensland University of Technology (QUT). (Not publicly available).
16. Cramb, S. M., E. W. Duncan, K. L. Mengersen, and P. D. Baade.  2018.  Australian Cancer Atlas, small-area survival: technical report.  Brisbane: Cancer Council Queensland and Queensland University of Technology (QUT).  (Not publicly available).
17. Cramb, S. M., E. W. Duncan, P. D. Baade, and K. L. Mengersen.  2018.  Investigation of Bayesian spatial models.  Brisbane: Cancer Council Queensland and Queensland University of Technology (QUT).  URL: https://eprints.qut.edu.au/115590.
18. Duncan, E. W., N. M. White, and K. Mengersen.  2017.  Spatial smoothing in Bayesian models: a comparison of weights matrix specifications and their impact on inference.  International Journal of Health Geographics 16 (47).  DOI: 10.1186/s12942-017-0120-x.
19. Cramb, S. M., E. W. Duncan, N. M. White, P. D. Baade, and K. L. Mengersen.  2016. Spatial Modelling Methods. Brisbane: Cancer Council Queensland and Queensland University of Technology (QUT).  URL: https://eprints.qut.edu.au/204103.
20. Duncan, E. W., N. M. White, and K. Mengersen.  2016.  Bayesian spatiotemporal modelling for identifying unusual and unstable trends in mammography utilisation. BMJ Open 6 (5): p.e010253. DOI: 10.1136/bmjopen-2015-010253.
21. Pokorny, M. R., M. de Rooij, E. Duncan, F. H. Schröder, R. Parkinson, J. O. Barentsz, and L. C. Thompson. 2014.  Prospective Study of Diagnostic Accuracy Comparing Prostate Cancer Detection by Transrectal Ultrasound–Guided Biopsy Versus Magnetic Resonance (MR) Imaging with Subsequent MR-guided Biopsy in Men Without Previous Prostate Biopsies. European Urology 66 (1): 22-29.  DOI: 10.1016/j.eururo.2014.03.002.

Under review/ in progress:

1. Kobakian, S. R., D. Cook, and E. Duncan.  A hexagon tile map algorithm for displaying spatial data.  In preparation for submission to The R Journal.
2. Kobakian, S. R., D. Cook, and E. Duncan.  Comparing the effectiveness of the Choropleth map with a hexagon tile map for communicating cancer statistics.  In preparation for submission to TBA.

Supervised Dissertations

Kobakian, S. R. 2020. New algorithms for effectively visualising Australian spatio-temporal disease data. Master of Philosophy thesis, Queensland University of Technology. URL: https://eprints.qut.edu.au/203908/1/Stephanie_Kobakian_Thesis.pdf

Mentoring and special mentions:

Aswi. 2020. Bayesian spatio-temporal modelling of small areas: Dengue fever in Makassar Indonesia. PhD Thesis, Queensland University of Technology. URL: https://eprints.qut.edu.au/200547/1/_Aswi_Thesis.pdf.

Jahan, F. 2021. New insights into Bayesian models for spatial data.  PhD Thesis, Queensland University of Technology. URL: https://eprints.qut.edu.au/212622/1/Farzana Jahan Thesis.pdf.

Guides

• A guide to installing R, RStudio, R Packages, and related software: RPubs link.
• A gentle introduction to R Shiny apps: shinyapps.io link.
• Publicly available spatial data sets for health research: RPubs link.
• Creating spatial polygon objects and how to manipulate them: RPubs link.
• Creating artificial maps and shapefiles in R: RPubs link (v1.0 currently only includes square lattices).
• Computing centroids of polygons – bounding box vs geometric centre methods, pros and cons: RPubs link.
• A guide to working with MySQL and SQL Server via R: A data scientists’ dream pipeline (Coming soon).

Link to Earl’s RPubs website: http://rpubs.com/Earlien.

BRAG Talks

• 18 Aug 2016: Deriving the Full Conditionals (updated version here).
• 20 July 2017: Reversing Label Switching.
• 29 March 2018: Cartograms.
• 17 Jan 2019: Attempts to Quantify Under- and Over-smoothing (see the slides from the Newcastle University invited talk, 4 Sept 2019, for a more polished version; see also the Bayes on the Beach 2019 poster presentation).
• 7 Nov 2019: MCMC Sampling: creating a generic sampler & comparison of MCMC algorithms (revised 9 January 2020).

Invited Talks

• 28 Aug 2019, GeoMed Conference, Glasgow: The Australian Cancer Atlas: mapping reliable small-area estimates of cancer incidence and survival.
• 4 Sept 2019, Newcastle University Talk, UK: Comparing spatial models in the presence of spatial smoothing (also presented at the Alan Turing Institute, London (23 Aug), University of Glasgow (30 Aug), and University of Edinburgh (2 Sept)).  This work has now been published in PLOS ONE.
• 22 Nov 2019, OzViz Conference, Brisbane: Novel visualisations for a digital, interactive cancer atlas.
• 16 Dec 2019, Online presentation for Western Australian Dept. of Health (WADoH): Exploring the Australian Cancer Atlas: journey into (previously) uncharted territory.

Poster Presentations

ACEMS Retreat 2016, Gold Coast

Bayes on the Beach 2019, Gold Coast

Peer Review

I am a regular reviewer for the International Journal Health Geographics (IJHG), International Journal of Environmental Research and Public Health (IJERPH), and International Journal of Geo-Information (IJGI) amongst other journals.

For a full list of verified peer reviews, please visit my Publons profile.

Awards and Special Recognition

• For the Australian Cancer Atlas:
  • 18 Oct 2019: Asia-Pacific Spatial Excellence Awards for Spatial Enablement (regional Queensland Winner)
  • 28 May 2020: Winner of the 2019 Oceanic APSEA Award for Spatial Enablement
  • 28 May 2020: JK Barrie Award for Overall Excellence
• 31 Oct 2019: ACEMS Impact and Engagement Award (awarded to Farzana Jahan, Kerrie Mengersen, Earl Duncan, Susanna Cramb, Dianne Cook, Stephanie Kobakian, and Nicole White).

Contact Details

Email: earl.w.duncan@gmail.com

Other links:

• LinkedIn profile
• Google scholar profile
• ACEMS profile
• QUT profile