About CompOnc


Computational Oncology at the Cavendish Laboratory

The Cavendish High Energy Physics group are contributing to the Large Hadron Collider experiments at CERN. Incredible quantities of data are generated from these experiments, generally as images of the particle collisions. HEP people have the knowledge and experience in dealing with these large volumes of data, in terms of using advanced batch processing, categorising and storing techniques. On the other hand, advanced radiotherapy systems also generate comparatively large quantities of medical images that have to be categorised and stored, as well as processed. Radiotherapy is an essential aspect of cancer treatment, responsible for destroying the cancer in 40% of those patients who are cured. It makes a greater contribution to overall survival than chemotherapy.

In radiotherapy, the total prescribed radiation dose is often divided onto a number of daily sessions, called fractions, which are delivered over few weeks. Treatment is usually delivered with Intensity Modulated Radiotherapy (IMRT), a technique to treat complex shapes, based on sophisticated computation. Treatment is planned based on findings obtained by investigations using diagnostic images such as CT and MRI scans. Optimal parameters of radiotherapy are selected at the planning stage in order to accurately target the tumour with radiation but to spare surrounding healthy tissues. This mainly includes shaping and tweaking the radiation beams in order to match the individual tumour’s characteristics.

Tumours usually change in shape, size and location during the course of treatment due to organ movements. Image-Guided Radiotherapy (IGRT), a technique that extensively uses medical imaging, can track the position of the target before each treatment, ensuring the dose is delivered accurately. The treatment parameters are updated according to the findings obtained using IGRT. However, current methods involve labour-intensive and manual operations, which can render errors in the treatment process. These factors introduce a bottleneck in the clinical delivery workflow, remarkably with the increasing number of cancer patients, which can cause long queues and less survivability rates.

We employ novel computational methods aiming to speed up the whole treatment process and prevent excessive tissue damage. Our research draws on the strengths of a long standing and fruitful collaboration between the Departments of Physics and Oncology in the study of computational radiotherapy. Starting in 2011 with the AccelRT project, followed by VoxTox, the collaboration has also been successful in securing funds for several PhD students, a Wellcome Foundation internship for the GHOST project and two EPSRC/STFC Impact Acceleration Account awards to explore the commercial potential of its work. Our mature multidisciplinary research group builds on existing expertise in imaging, high energy physics, and high performance computing in the Cavendish Laboratory, together with experience in clinical radiotherapy and imaging in the Oncology Centre at Addenbrooke’s Hospital.