One of the unwritten parts of the job description of a trainee neurologist working in an academic institution can be the requirement to enrol into a brain MRI study as a research volunteer (‘normal control’ sounds too flattering for a neurologist). During my training, thrilled by the respite from requesting esoteric blood tests, I gladly lay on the research MRI scanner table and imagined my protons spinning under the influence of the magnet around me. But had I known that there was a 1 in 37 chance of identifying an incidental finding on brain MRI,1 the magnet would have seemed more like a European roulette wheel in which the ball may land in one of 37 pockets (gamblers, scholars and pedants who read Practical Neurology will know that this comparison breaks down in American roulette where the existence of a 38th double zero pocket offers the player longer odds).

However, before deciding whether to be concerned about the chance, and perhaps the risk, of detecting incidental findings in the brain, it is important to clarify what they may be. In a meta-analysis of their prevalence in the brains of neurologically asymptomatic people, my colleagues and I eventually defined them as ‘apparently asymptomatic intracranial abnormalities that are clinically significant because of their potential to cause symptoms or influence treatment’ (this …