A biomarker is defined as a measured characteristic that is an indicator of normal biological processes, pathogenic processes or response to an exposure or intervention, including therapeutic interventions. The current FDA/NIH definition states explicitly that molecular, histologic, radiographic or physiologic characteristics are examples of biomarkers. Magnetic Resonance Imaging has unparalleled potential to generate imaging biomarkers from intrinsic differences in tissue relaxation properties, fat content, water diffusion, vascularity, elasticity and characterization of the molecular environment of MR visible atoms. In addition, the injection of extrinsic weakly paramagnetic agents such as gadolinium chelates further determines tissue vascular state by generating semi-quantitative and quantitative parameters relating to tissue perfusion and permeability from dynamic contrast-enhanced (DCE) techniques. This information on tissue structure and/or function can be used qualitatively and quantitatively to assess disease status, therefore potentially providing invaluable imaging biomarkers. To be of use in clinical decision-making, biomarker(s) must improve disease detection, aid staging or provide prognostic information or robust response assessment and follow-up. Disease detection and staging are usually done by qualitative, subjective assessment of images, whereas prognostic or response assessment biomarkers require quantitative evaluation. To become clinical decision-making tools for use in healthcare either as companion diagnostics, or as screening, prognostic, predictive or response biomarkers, imaging biomarkers require validation and qualification in multicentre trials.The process of validation and qualification of an Imaging Biomarker requires 2 steps. The first examines technical performance and relates to biomarker precision; the second examines biological performance and relates the biomarker to a specific aspect of biology or outcome. Unfortunately, the variety of acquisition and analysis methodologies available in generating the data have meant that MR biomarkers remain non-standardized in both acquisition and analysis. This leads to variability of the data, invalidates their implementation in multicentre clinical trials and makes accrual of meaningful outcome data difficult. Recently, there has been an international move to address this problem. The questions of how the acquisition and analysis of various Imaging Biomarkers should be standardized and how terminology should be harmonized have been addressed by numerous representatives from academia, clinical medicine, industry and the regulators. Collaborative working between North America’s, Quantitative Imaging Biomarkers Alliance (QIBA) and Quantitative Imaging Network (QIN) and the European Imaging Biomarkers Alliance (EIBALL) with input from the American College of Radiology Imaging Network (ACRIN), the European Society of Radiology (ESR), European Organization for Research and Treatment in Cancer (EORTC), the International Society for Magnetic Resonance in Medicine and Cancer Research UK is leading to the development of standards for quality assurance and quality control of imaging within clinical trials through use of standardized test objects, standardized protocols and readouts.Failure to integrate an MRI biomarker into radiology practice may also occur for reasons other than lack of technical standardization. Left ventricular ejection fraction (LVEF) is an example of a highly successful volumetric measurement derived from end systolic and end diastolic volumes on cardiac MRI. Cardiac MRI is regarded as the gold standard method for making measurements of LVEF because of its high reproducibility and is fully integrated into radiology practice in all cardiac centers. Tumor volume on the other hand, which could be derived just as quickly and easily is rarely used. This is because a simple unidimensional measurement correlates sufficiently well with outcome making entire volume segmentation unnecessary. Similarly, simple qualitative assessment using a single functional metric (temporal pattern of DCE) in conjunction with morphological imaging is standard practice in the assessment of breast masses, while a combination of qualitative assessments (apparent diffusion coefficient [ADC], T2-W appearance and temporal pattern of DCE) is used routinely to assess the probability of malignancy within the prostate (PI-RADS v2). Both these qualitative assessments are easily incorporated into the clinical workflow and have been biologically validated as a sufficiently accurate readout of the presence of malignancy. Conversely, even extremely promising biomarkers such as the ADC are not used quantitatively as prognostic or response biomarkers because of the hitherto variability of the measurement across different hardware and software platforms. This is currently being addressed through European Union initiatives such as the QuIC-ConCePT project funded by the innovative Medicines initiative, which has enabled multicentre standardization of this biomarker in oncology trials. As MRI matures in its capability and becomes ever more widely available, we are moving to an era where biomarker discovery needs to move swiftly through a process of standardization, validation and qualification in order to implement useful MRI biomarkers into routine clinical care.