Don't let the data cloud the image

Abstract

It is no surprise that many radiologists feel overwhelmed with the large amount of data they have to examine. It was estimated that all the medical images captured in the USA in 2014 amounted to 100 petabytes of information, or the equivalent of 223 000 DVDs, and the global total is forecast to hit 35 zettabytes or 1 million petabytes by 2020. With the advent of personalised medicine, the processing power needed to handle the expansion of imging techniques such as CT, PET, and MRI looks set to surge even further. The audience at 42 Ways to See a Tumour, one of the events at the 2018 Edinburgh International Science Festival, could have been forgiven for suffering from information overload too. Michael Jackson, consultant radiologist (Royal Hospital for Sick Children, Edinburgh, UK), began his talk by bombarding the crowd with 42 images of a rare tumour found in a 12-year-old patient who arrived at the hospital just over a year ago. The patient had had three weeks of blurred vision caused by extraordinarily high blood pressure, which can cause changes at the macula, the part of the eye responsible for central vision. Diagnosis began with an ultrasound scan of the patient's liver and, although Jackson was confident in his diagnosis after the ultrasound, he showed how he used MRI to confirm his suspicions. The MRI scans revealed a pheochromocytoma, a liver tumour that releases hormones that can raise the patient's blood pressure, which probably caused the damage to the macula.