A plug-in to extend clinical imaging quality control to pixel data, using computer vision techniques to characterise the images. Specific QC tasks include checking for “burned-in” annotation (that often contains personally-identifiable information) and anatomical region/completeness (e.g. to label the images as head, chest, abdomen, or pelvis; or label organs).