How can artificial intelligence (AI) support both cancer research and healthcare more generally? Anoushka Harit (University of Cambridge) and Dr Zhongtian Sun (University of Kent) join today's episode to discuss this topic.

Anoushka talks about her research in hereditary diffuse gastric cancer (HDGC) and the development of an AI system to enable early detection of signet ring cell carcinoma (SRCC) during endoscopy. This new AI system could help to treat SRCC better or to prevent or delay drastic life-changing treatment, preserving patients' quality of life.

Zhongtian shares insights on how AI can support health applications more generally and how he tries to make it as explainable as possible. He even talks about his wish to make the AI mimic as much as possible the way a human brain works!

Integrated Cancer Medicine: Research in Focus

Using AI to detect stomach cancer during endoscopy

Have you ever wondered how artificial intelligence (AI) is used in research in a healthcare setting? Rebecca Wray and Bill McGough, two PhD students at the University of Cambridge, both use AI in different ways in their research into the screening and treatment of kidney cancer.

Rebecca is an MRes PhD student investigating the response and resistance to therapy in patients with kidney cancer. She uses AI to find tumour features at multiple scales. For his PhD, Bill is developing AI models that will sift through the mounds of imaging data produced during the Yorkshire Kidney Screening Trial (funded by Yorkshire Cancer Research), flagging images that show masses in the kidneys.

This interesting conversation gives insight into how researchers are using patient data safely in order for AI models to integrate all the information about a patient to help review their treatment and personalise it. The discussion also highlights how using AI in the screening setting is speeding up processes to help save clinicians' time and allowing the potential earlier detection of kidney cancer in a cost-effective manner.

Using AI in the screening and treatment of kidney cancer

Can we use artificial intelligence (AI) to predict where a brain tumour might recur at the point of first diagnosis in order to pre-emptively treat it to stop recurrence?
 
Dr Francesca Cozzi (a PhD student at the University of Cambridge) and Dr Curtis Holliman (Associate Professor in the Mathematics Department at The Catholic University of America, USA) believe this will eventually be possible.
 
In this exciting conversation they talk about their collaboration to find a mechanism to mathematically model the progression of glioblastoma. They highlight why it is important to achieve this in an explainable way using AI, to keep the process of decision making transparent and interpretable to self-validate the models they develop.

Using AI to mathematically model the progression of glioblastoma

The Cancer Research UK Cambridge Centre unites over 1,000 world-leading biologists, chemists, physicists, engineers, mathematicians, computer scientists, clinicians, nurses and allied healthcare professionals from across Cambridge, UK to tackle cancer from every angle.

Our mission is to end death and disease caused by cancer through research, treatment and education. We are detecting cancer at its earliest stage and are developing personalised treatments for every patient through facilitating new collaborations and driving the translation of new scientific discoveries into clinical applications to improve patient care. By working together across a range of different disciplines, our members are breaking down the barriers between the laboratory and the clinic, enabling patients to benefit from the latest innovations in cancer science.

New tracks tagged #University of Cambridge