Alberta Cancer Foundation

Embracing Artificial Intelligence in Cancer Research

Illustration by Breanne Beckie.

More and more these days, artificial intelligence (AI) applications are being embraced by organizations in a variety of sectors — and the cancer care industry is no exception.

AI is the simulation of human intelligence and decision-making by computers. “Machine learning” and “deep learning” are two subsets of AI that are often used in health care research. Each involves inputting a large amount of data into a computer. The AI analyzes this data and applies what it has learned to new data. Medical researchers around the globe are starting to take advantage of the real-world potential available in this swift analysis of large amounts of medical information.

Several examples of this can be found in London, England. There, an AI research company called DeepMind (a subsidiary of Google) has partnered with the radiotherapy department at University College London Hospitals NHS Foundation Trust to develop and test the use of AI in segmenting and analyzing medical images, such as CT scans, to help head and neck cancer patients undergo faster treatments. Initial findings, published in September 2018, confirmed that AI can analyze and segment scans as accurately — and much more quickly — than experienced radiologists can. Now, researchers are moving on to the next stage of testing, which involves assessing the AI’s performance in a clinical environment.

DeepMind is also involved in another cancer research project, in conjunction with London’s Cancer Research UK Imperial Centre and Tokyo’s Jikei University Hospital. Each partner will provide results from around 30,000 mammograms (with all personal information removed) to research the possibility of using AI to screen potential breast cancer diagnoses, with the aim of reducing missed or inaccurate findings in mammograms.

Another advance using AI has been developed by scientists at The Institute of Cancer Research, London and the University of Edinburgh. REVOLVER (Repeated evolution of cancer) analyzes vast amounts of data on DNA mutations and its patterns to help predict the behaviour and development of tumours. The hope is that it will allow doctors to create treatment plans for patients with new levels of precision, including the ability to predict future resistance to treatment or a change in tumours.

With the potential for faster diagnosis, more targeted and precise treatments and reduced data work for health care workers, the benefits of AI promise to be deeply felt in the future.

It is predicted that the next five years will bring more advancements in artificial intelligence than the last 30 years combined.