The Institute of Cancer Research, London, is one of the world’s most influential cancer research institutes, with an outstanding record of achievement dating back more than 100 years. We provided the first convincing evidence that DNA damage is the basic cause of cancer, laying the foundation for the now universally accepted idea that cancer is a genetic disease. Today, The Institute of Cancer Research (ICR) leads the world at isolating cancer-related genes and discovering new targeted drugs for personalised cancer treatment.
Under the leadership of our Chief Executive, Professor Paul Workman FRS, the ICR is ranked as the UK’s leading academic research centre. Together with our partner The Royal Marsden, we are rated in the top four cancer centres globally.
The ICR is committed to attracting, developing and retaining the best minds in the world to join us in our mission – to make the discoveries that defeat cancer.
We are seeking a highly motivated data scientist to perform state-of-the-art analyses of radiology image data, using radiomics, machine learning and other innovative methods. The successful candidate will work on data from a variety of different clinical trials and will contribute to the integration of images from different modalities with clinical data, “-omics” and other types of cancer data.
This two-year, fixed-term post within the Division of Radiotherapy and Imaging is an exciting opportunity for a scientist with a flair for statistics and coding, at the leading edge of informatics in cancer research: the interface of radiology and genomics.
You will be working on a programme whose overall aims are to apply advanced information- and image-processing techniques to tumour classification and early detection of response to treatment. You will be involved in the study of both rare and common cancers, which provide interesting and different challenges, the solutions of which can make a real difference to patients’ lives.
Radiology is a domain that could realise enormous benefits from Big Data and Artificial Intelligence. There is a clear capacity problem in the sector: our MRI and CT scanners can deliver data at a faster rate than it can be “read” by increasingly overstretched radiologists, whilst an aging population means that demand will continue to increase faster than new human practitioners can be trained. Computers perform in a consistent and reproducible manner, do not tire and can be replicated easily. It is well recognised that we are not currently making the most of the wealth of radiology data available. Simple image-based features, such as the longest diameter across a tumour in a 2-D image slice (part of the RECIST set of criteria) fail to capture the subtleties of tumour shape and image texture. Radiomics is the name given to a data-processing paradigm by which quantitative imaging “features” are extracted from pixel data (typically via automated algorithms) and analysed using statistical methods and machine learning.
The Institute of Cancer Research and Royal Marsden Hospital (ICR/RMH) have access to a unique collection of clinical research data encompassing a range of tumour types, including breast, lung, myeloma, prostate and sarcoma, and you will be responsible for developing and delivering radiomics-based analyses in a number of these areas. A particularly exciting data collection will be generated from the TRACERx – Renal study. TRACERx (TRAcking Cancer Evolution through therapy (Rx)) is a long-term, multi-million pound research project that will transform our understanding of the drivers and intratumour heterogeneity of renal cell carcinoma, and determine the relationship between heterogeneity and disease stage, clinical outcome and treatment response. This will be a huge step towards an era of precision medicine.
Recent work at the ICR has established a radiomics “pipeline”, based on the XNAT data-sharing platform (
www.xnat.org), which will rapidly deliver the quantitative “feature” data. The aim of this post is to perform the subsequent analysis and interpretation, building descriptive and predictive models to relate imaging findings to non-imaging biomarkers.
The successful candidate will have considerable relevant experience and will be expected to possess a relevant first degree in an appropriate scientific or computational discipline.
Appointment will be on a Fixed Term Contract for 24 months. The starting salary will be £31,023 – £38,135 p.a. inclusive, dependent on ability, experience and fit to the person specification.
All applications should be submitted online on the ICR Careers web page
https://www.icr.ac.uk/careers , vacancy ref 657
