Published on April 1st, 2016 | by Theo Arvanitis0
Radiomics: using the power of informatics to more accurately diagnose and treat childhood brain tumours
Professor Theo Arvanitis talks to BQ Magazine about using the power of informatics to accurately diagnose and treat childhood brain tumours. Theo Arvanitis is a Professor of e-Health Innovation and Head of Research at the Institute of Digital Healthcare, WMG, University of Warwick.
Although childhood cancer is rare – in the UK, statistics predict that approximately only one child in 500 will develop cancer by the time they reach 14 years of age – the disease poses a huge struggle for the affected children and their families. In particular, brain and other central nervous system tumours are the most common causes of death from cancer in children. Thanks to advances in diagnostic technologies, new treatments, and improvements in the clinical management of patients with the disease, we have seen considerable improvement in survival rates from childhood brain tumours in recent years. However, the necessity for non-invasive diagnostic and prognostic characterisation of childhood brain tumours still poses a considerable challenge. We need to understand the disease in order to predict its progression and manage it effectively, but we cannot simply open up the brain to do this; which is where the role of medical imaging becomes so important.
Magnetic resonance imaging (MRI) is the key technology used for the visualisation of the presence of cancer in the brain. It can support medical staff in the precise diagnosis of the tumour type, and enable them to make informed decisions on the management of the disease. The effective clinical management of childhood brain cancer patients is complex and time-consuming, and so the combination of this imaging with new informatics technologies (e.g., computer processing, modelling, algorithmics) can provide powerful tools in the hands of oncologists.
Radiomics is an emerging and exciting field involving conversion of the images from standard medical scans into data that can be mined and processed to identify useful characteristics, patterns, and disease markers, including for example MRI image texture. Gathering this data into one place and running it through complex software, incorporating computational and machine learning techniques, will not only aid tumour diagnosis, but provide more accurate prediction of disease progression, responsivity to treatments, and patients’ survival. This not only greatly improves the potential to advance clinical management of patients for successful therapy and cure, but supports more informed discussions between medical staff, patients, and their family on how to manage the challenges of the disease. As a consequence, MRI texture analysis is emerging as a powerful weapon in the arsenal of biomedical technologies for us to win the battle against childhood cancer in the brain and central nervous system.
At the Institute of Digital Healthcare (IDH), in collaboration with paediatric neuro-oncologists at Birmingham Children’s Hospital, we are investigating the application of such informatics and advanced computer algorithms to provide an intelligent decision support aid to clinicians for the characterisation of childhood brain tumours. We believe that the impact of digital technology and information can be transformative for not only healthcare, but to individuals and society. Children’s health journeys can be better understood and, as a consequence, we can reduce the burden of the disease for affected children and their families, while making further steady steps in the battle against cancer.