Development of imaging biomarkers and generation of big data.

Several image processing algorithms have emerged to cover unmet clinical needs but their application to radiological routine with a clear clinical impact is still not straightforward. Moving from local to big infrastructures, such as Medical Imaging Biobanks (millions of studies), or even more, Federations of Medical Imaging Biobanks (in some cases totaling to hundreds of millions of studies) require the integration of automated pipelines for fast analysis of pooled data to extract clinically relevant conclusions, not uniquely linked to medical imaging, but in combination to other information such as genetic profiling. A general strategy for the development of imaging biomarkers and their integration in the cloud for the quantitative management and exploitation in large databases is herein presented. The proposed platform has been successfully launched and is being validated nowadays among the early adopters' community of radiologists, clinicians, and medical imaging researchers.

Current knowledge on tumour induction by computed tomography should be carefully used.

Risks associated to ionising radiation from medical imaging techniques have focused the attention of the medical society and general population. This risk is aimed to determine the probability that a tumour is induced as a result of a computed tomography (CT) examination since it makes nowadays the biggest contribution to the collective dose. Several models of cancer induction have been reported in the literature, with diametrically different implications. This article reviews those models, focusing on the ones used by the scientific community to estimate CT detriments. Current estimates of the probability that a CT examination induces cancer are reported, highlighting its low magnitude (near the background level) and large sources of uncertainty. From this objective review, it is concluded that epidemiological data with more accurate dosimetric estimates are needed. Prediction of the number of tumours that will be induced in population exposed to ionising radiation should be avoided or, if given, it should be accompanied by a realistic evaluation of its uncertainty and of the advantages of CTs. Otherwise they may have a negative impact in both the medical community and the patients. Reducing doses even more is not justified if that compromises clinical image quality in a necessary investigation. Key Points • Predictions of radiation-induced cancer should be discussed alongside benefits of imaging. • Estimates of induced cancers have noticeable uncertainties that should always be highlighted. • There is controversy about the acceptance of the linear no-threshold model. • Estimated extra risks of cancer are close to the background level. • Patients should not be alarmed by potential cancer induction by CT examinations.