Information quality, normalization and data safety analysis

Overview

The poster discusses label correctness, simulated-data realism and normalization of colour/contrast and camera parameters. Its iterative data safety analysis links test or monitoring findings to limiting data features, then distinguishes updates to datasets, models/monitors or requirements according to the identified cause.