How to get started with AI in oncology
Article by: Professor Raj Jena
Professor Raj Jena, Clinical Professor of Data Science and Machine Learning in Radiotherapy at University of Cambridge, and the RCR’s newly appointed Clinical Oncology Lead AI Advisor, explores the real-world impact of AI in oncology – and what residents can do now to feel more confident working with AI.
You’re responsible for one of the streams at the RCR’s Global AI Conference 2026 – could you tell us about your session and what delegates will gain from attending?
I’m proud to be working with the team on stream two – Future Ready AI. We are looking forward to talks highlighting new innovations in quantum computing to power the next generation of AI applications, sustainability in AI and understanding how AI is working behind the scenes in your working life. We also have a very exciting programme on AI in theranostics, bringing together the two specialties of our College.
Where do you see AI currently making the most difference to patient outcomes in clinical oncology?
Right now, AI is primarily a time-saving and quality assurance tool. We use it to automate treatment planning and enhance the quality of radiotherapy. We’ve been successful in this regard, but I, for one, am looking forward to more transformative breakthroughs in clinical oncology, such as AI enabled MDTs and AI powered adaptive radiotherapy.
What’s one AI related skill you think every resident should start developing now?
It’s important to have a go at coding an AI model using Python. You don’t need a lot of knowledge, but some experience of writing a script that uses an AI framework to help you solve a task is so helpful when trying to understand how larger-scale AI algorithms work.
What’s one thing residents can do now to feel more confident engaging with AI in their day-to-day work?
Doing a course – ideally one that gives you some kind of credential – can really help you get over the initial activation barrier to working with AI. There are many online courses available (see the College’s AI fundamentals for imaging and healthcare). The key is to think laterally; it doesn’t have to be an oncology application or even a medical one.
What’s one common misconception about AI in oncology you’d like to challenge?
Many people believe AI will replace oncologists. No technology in the last 100 years has replaced us – it simply augments our capability to delivery precision cancer care.
It’s also my observation that the cutting-edge AI exemplars in healthcare that make use of foundation models or LLMs are just that: exemplars. They need much more work to get them to clinical utility.