Skip to contents

Changelog

Source: NEWS.md

epidscan 0.1.0

First release of epidscan - A modern, stateless R interface for SaTScan spatial scan statistics.

Features

Core Analysis Engine

  • satscanr() - Stateless analysis engine that accepts validated inputs and returns parsed results directly
  • Automatic parameter management - Three-level hierarchy (data defaults → function overrides → template files) for flexible configuration
  • Clean workspace - Results generated in tempdir() by default; use output_dir to save explicitly

Type-Safe Input System

The ss_tbl system ensures data is correct before analysis runs:

  • ss_cas() - Create validated case files with upfront error checking
    • Zero-case handling
    • Time formatting with precision control (day/month/year/generic)
    • Support for covariates (age groups, sex, etc.)
  • ss_geo() - Spatial coordinates with native sf support
    • Automatic centroid extraction from polygons
    • CRS detection (geodetic vs. cartesian)
    • Correct coordinate ordering for SaTScan
  • ss_pop() - Population denominators
    • Zero-population checks
    • Time-based population with precision control
    • Covariate support matching case files
  • ss_ctl() - Controls for Bernoulli models
  • ss_grd() - Custom scan centers

Parameter Management

  • prm_options() - Object-oriented parameter configuration for reproducible analyses
  • prm_parse() / prm_write() - Robust parser that preserves comments and structure
  • prm_validate() - Validate settings against official SaTScan templates before execution

Results & Visualization

  • S3 methods following broom conventions:
    • tidy() - Cluster-level details as tidy data frames
    • glance() - Analysis-level summary metrics
    • summary() - Human-readable cluster summaries
  • map_clusters() - Interactive visualization with leaflet
    • Significance-based color coding
    • Rich cluster popups with statistics
    • Support for both cartesian and lat/long data

Smart Defaults

  • Automatic date inference - Time range detected from your data if not specified
  • Flexible column handling - Works with your column names via role mapping
  • Spatial-first - Seamless integration with sf simple features

Included Data

  • New Mexico lung cancer dataset - Complete example with cases, population, and coordinates
    • NMlung_cas - Monthly case counts by county, age, and sex
    • NMlung_pop - Annual population by county, age, and sex
    • NMlung_geo - County centroids in kilometers

Documentation

  • Getting Started vignette - Complete workflow walkthrough
  • Parameter Reference - Guide to SaTScan parameter configuration
  • Function documentation - Comprehensive help pages for all functions

Requirements