aceBuilding and Validating a Data-Governance Workflow for Artificial Intelligence-Based Tuberculosis Detection from Chest Radiographs, and Quantifying the Impact of Data Leakage
Studies applying artificial intelligence (AI) to tuberculosis (TB) detection from chest radiographs (CXR) focus on models and metrics, yet the quality, provenance and reproducibility of the data workflow decide the reliability of results; duplicate images and non-patient-level splitting inflate evaluation metrics. We propose a seven-step data-governance workflow: multi-source standardisation, provenance management, three-hash deduplication (MD5, pHash, dHash), patient-level splitting, invariant-check auditing, l…
A seven-step data-governance workflow cleaned and sealed a 46,410-image chest radiograph corpus and made leakage effects verifiable and measurable for AI tuberculosis research.
Random image-level splitting inflated internal AUC on a single-source drug-resistance task by 1.7 percentage points, so leakage can overstate how reliable AI tuberculosis models look.
The saturated AUC near 0.997 across scenarios is consistent with source-class confounding, so the multi-source task could not show leakage effects well, and a deduplication gap was found only during revision.
Evidence
- Peer-reviewedJournal of Imaging Informatics in Medicine2026-10-16
How should this claim be treated?
Truvace Impact Record TRV-2026-1930, v1: “Building and Validating a Data-Governance Workflow for Artificial Intelligence-Based Tuberculosis Detection from Chest Radiographs, and Quantifying the Impact of Data Leakage.” Truvace, 2026-10-19. /record/TRV-2026-1930 (accessed at citation time). sha256 56e3f15045417e85…
Calibration history
Every change to this record since certification, in the open. None yet — the reading has held since it entered the record.
Certified into the record
How to verify without trusting this page
Fetch the canonical text of any version from /api/record/TRV-2026-1930 and hash it yourself — for example shasum -a 256 on the saved canonical field. The result must equal content_hash, and each version’s text ends with prev:followed by the prior version’s hash (version 1 chains to 64 zeros). If a single character of any version had been altered since certification, the chain would not reproduce.