Mapping the spread of cancer with deep learning-powered whole-body magnetic resonance imaging

Project ID: 2531ad1523

(You will need this ID for your application)

UCL Lead department: Division of Medicine

Project Summary:

In this project you will develop deep learning-powered magnetic resonance imaging (MRI) techniques to aid in treatment decision-making in patients with cancer which has spread to multiple regions in the body.

Cancer, which is uncontrolled proliferation of the body’s cells, can harm the body beyond the tissue in which it formed. This spread of cancer, known as metastasis, is caused by cancer cells breaking away from their original site and being transported through the blood or lymphatic system to other parts of the body, forming new tumours. Managing metastases is a key part of routine cancer care.

Doctors do now have powerful treatments available for treating metastatic cancer, but these treatments need to be given at the right time and have harmful side effects, meaning that it is crucial to use them sparingly and in a targeted way. This presents a problem for doctors – how to accurately determine how far the cancer has spread, and thus choose the right treatment.

Recently, whole body-MRI (WB-MRI) has emerged as a promising method for mapping the spread of cancer. However, current WB-MRI techniques are prone to confusing healthy tissue with cancerous tissue, and are slow, meaning that patients who are in pain need to lie in the scanner for around an hour per scan. In this project, you will develop new deep learning-powered WB-MRI methods that enable more accurate distinction between these tissue types, with reduced scan time, thus reducing discomfort for patients and helping doctors get the information they need to delivered targeted, effective cancer treatments.