Artificial intelligence is being used to predict cancer outcomes and survival

Southampton researchers are using AI to predict treatment outcomes and survival from advanced eye cancer that has spread to the liver.

The new study is using machine learning with a clinical data of metastatic uveal melanoma.

The insight could show which patients are most likely to benefit from treatment.

Scientists from the University of Southampton and University Hospital Southampton (UHS) are collaborating on the work, alongside experts from the Netherlands.

Spotting clues to survival outcomes

Uveal melanoma is the most common primary eye tumour in adults but has a very poor prognosis once the tumour has spread to the liver (liver metastases), with only one in 10 patients alive after one year.

University Hospital Southampton specialises in providing a promising minimally invasive treatment for liver metastases called Melphalan Percutaneous Hepatic Perfusion.

This involves delivering chemotherapy directly to the blood supply of the liver metastases via a small hole in the groin and neck. This has shown great promise, but outcomes vary significantly.

Dr Ganesh Vigneswaran, study lead, said: “While cancer treatments are helping many people, we cannot predict who will respond well to treatment.

“Machine learning is a type of AI that can find complicated patterns in big datasets, such as treatment response or survival outcomes. Most patients have routine CT scans and these might contain important clues specific to each person.

“Our goal is to employ AI to extract and uncover this information to help predict treatment responses. If we can establish which patients are likely to respond and which treatments are likely to be useful, we can improve decision-making and save patients from ineffective treatment and side effects.”

Foundation for other cancers

The research is funded by an Academy of Medical Sciences Starter Grant for Clinical Lecturers.

Team members plan to develop a tool to support shared decision-making for metastatic uveal melanoma patients to improve current treatment paths.

These approaches might be transferrable to other cancers that are treated via the blood vessels, such as TACE (trans-arterial-chemoembolization) and SIRT (Selective-internal-radiation-therapy) and could serve as the foundation for the use of routine imaging in predicting patient outcomes in many other cancer types.

This project is a collaborative effort between multiple fields including, UHS Interventional Radiologists (Dr Sachin Modi, Dr Brian Stedman, Dr David Breen, and Dr Drew Maclean), UHS Oncologists (Dr Ioannis Karydis and Dr Matthew Wheater), as well as experts from the University of Southampton with Prof Tim Underwood and Prof John Primrose in Cancer Sciences and computer scientist Prof Mahesan Niranjan.

Interventional Radiologist Dr Mark Burgmans will also providing expertise from Leiden University Medical Centre in the Netherlands.