Spatial Epidemiology and Applied Data Science

DMgtAsia · Second Module (Intermediate level)

Authors
Affiliations

Prof. (Dr.) Biju Soman

AMCHSS, SCTIMST, Trivandrum

Dr. Arun Mitra

AIIMS Bibinagar

Published

June 26, 2026

Welcome

This is the Second Module (Intermediate level) of the DMgtAsia course. It covers spatial epidemiology and spatial data science in R, from GIS concepts through to disease-cluster detection on Indian public health datasets.

The material draws on spatial data science trainings and workshops run across India, adapted for the DMgtAsia course.

What you will be able to do

  • Explain GIS concepts (CRS and projections, spatial dependence, spatial autocorrelation, Tobler’s First Law) and how they apply to public health.
  • Run spatial exploratory data analysis and apply CDC cartographic guidelines to public health maps.
  • Read, transform and join spatial data with sf; extract raster values; geocode addresses; and build choropleths.
  • Run and interpret global and local spatial autocorrelation (Moran’s I, LISA), and map SIR/SMR to find disease clusters.

How the module is organised

The module has four sessions. Each pairs a lecture with hands-on R exercises and a quiz.

Session Theme Lectures Hands-on
Session-I Foundations of spatial thinking Intro to GIS · Concepts in SDS Reflection
Session-II Exploratory Spatial Data Analysis (ESDA) Exploring data with R · Visualization principles EDA on WHO TB data
Session-III Introduction to sf The sf package Exercises 1–4
Session-IV Spatial Data Analysis & Clustering (lab-led) Exercises 6–7 · John Snow · SIR/SMR · Capstone

For the full session list and run order, see the navigation bar, or go to Setup to prepare your machine.

Delivery

  • Self-paced: 4 weeks, one session per week.
  • Intensive workshop: 2 days. Day 1 covers Sessions I–II, Day 2 covers Sessions III–IV.

Acknowledgements

Developed at the Achutha Menon Centre for Health Science Studies, SCTIMST, and AIIMS Bibinagar. Built on the sf, spdep, rgeoda, ggspatial and tidyverse packages. Data: NMC, NFHS, WHO, SEDAC GPWv4, and HistData.