Skip to content
All projects
Research programmeLead Data Manager

Image Quality Study — X-ray Machine Standardisation

BCHPR · multi-country (Cameroon + Nigeria) · 2024 – present

5,173-line multi-country study comparing digital X-ray machine image quality (PRORAD vs Min Xray vs Fuji X-air) with DICOM metadata extraction and AI score differencing — informs equipment procurement and WHO guidance on digital X-ray adoption.

Highlights

  • Targeted DICOM extraction via pydicom across 43 predefined tags — supports custom-tag, all-tag, and protocol-default modes with multi-path source support.
  • Machine-to-machine AI score comparisons: prorad_minus_contemp_diff, min_prorad_diff, prorad_fuji_diff.
  • 9 REDCap reports merged via MultiFrameMerger (suffix map: dem · sym · his · xd1 · xd2 · sds · ia1 · ia2).
  • Country-specific study-ID patterns — Nigeria JHFIQS#### vs Cameroon CMIQS-#####.
  • Age-stratified BMI data-quality rules (adults 12–50, children 11–35).
  • Incremental DICOM processing with change detection — no reprocessing of already-ingested files.
  • Form-level user activity tracking with LRU-cached value cleaning for performance on thousands of events.
  • QueryBuilder-based DQ rules for AI score null detection, BMI validation, and duplicate-ID flagging.