Colour maps for representation of quantitative metrics in radiology: meaningful or misleading?

Abstract The increased utility of quantitative metrics in medical imaging has prompted the use of colour representation of the data. Colour schemes for image display can highlight pathology and distinguish normal anatomy, thus potentially aiding image interpretation. Adding colour schemes to images may serve the sole purpose of visualising an abnormality or may seek to represent and meaningfully convey a quantitative metric. An early use of colour representation is exemplified by colour Doppler in ultrasound (US), where a two-sided linear scale is commonly applied. More recently, many schemes have arisen, ranging from a single colour to depict a segmented region to a range of colours that embrace the entire colour spectrum. These may be applied as continuous or discrete palettes, depending on the software vendor or even on individual user settings. The lack of standardisation of colour schemes across the medical community, especially in medical imaging, risks misinterpretation of quantitative data and potential unintended negative clinical consequences. The purpose of this article is to review the colour schemes used for quantitative imaging across US, computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET), the way in which they are applied, the clinical applications where their use is desirable for enhanced interpretation, and the pitfalls where standardisation is lacking in a variety of clinical applications. Finally, we discuss some solutions that could be employed to enable standardisation and augment the use of colour in image interpretation. Key Points Question Colour schemes in medical imaging can highlight pathology, but standardisation is lacking for most imaging modalities, thus risking misinterpretation of quantitative data. Findings For T1, T2, and T2* maps and their inverses, colour map recommendations have recently been proposed, but are mostly lacking for other quantitative data. Critical relevance statement Standardised colour maps can enhance data analyses across scanners and institutions, which is important for multicentre studies and handling big data. Furthermore, standardised and harmonised colour maps could improve lesion conspicuity, cross-platform comparability, and longitudinal reliability.