Clatterbridge co-develops AI-based tool for specialist cervical cancer treatment

The Clatterbridge Cancer Centre in Liverpool has partnered with Canadian software company Limbus AI to trial and co-develop a commercial AI-based autocontouring tool for brachytherapy cervical cancer treatment.

Brachytherapy involves placing an intense radiation source inside or in close contact with tumour cells. This makes it well suited to treating tumours in particular parts of the body such as the cervix.

Although it is highly effective, some patients can find cervical brachytherapy uncomfortable as it involves lying flat for several hours with an applicator in place so clinical teams and physicists can plan and deliver the treatment. Using an AI-based tool to reduce the length of time this takes could significantly benefit patients as well as the professional teams involved. These tools have been used for some years in traditional radiotherapy planning, but until now there has been nothing similar available for cervix brachytherapy patients.

Clinicians and clinical scientists at The Clatterbridge Cancer Centre approached contacts at Limbus AI, a Radformation company, to suggest collaborating on a solution. The two organisations have worked together to develop and test what is said to be the world’s first commercial AI-based autocontouring software modules for cervix brachy-therapy. The tool has been successfully trialled at The Clatterbridge Cancer Centre and other hospitals and has just been launched commercially in the UK for use in clinical practice. It works by recognising and automatically outlining important organs in the body on MRI scans.

Dr Mona Chopra
Clinical oncologist Dr Mona Chopra using the autocontouring tool.

The day procedure begins with the patient having an applicator inserted in theatre under general anaesthetic. Once recovered, the patient has an MRI scan that creates a slice-by-slice view of the cancer and the vital organs around it so that clinicians and clinical scientists can create the treatment plan. Radiation treatment is delivered through the applicator and this is then removed.

Early clinical experience of the AI-based tool at The Clatterbridge Cancer Centre indicates an average time saving of around 13 minutes per treatment plan. As every patient needs three brachytherapy treatments, this could represent a significant improvement in patient experience.

Principal clinical scientist Rhydian Caines said: “As a clinical team we are keenly aware that brachytherapy can be a difficult experience for patients and were frustrated that we couldn’t find an autocontouring solution for brachytherapy, despite this technology being available in more traditional radiotherapy settings.

“The team at Limbus AI has been fantastic in answering our call to action on this, and it has been a thrill to  collaborate with them on bringing this system into existence. It’s early days and we’ll be continuing to work with Limbus on refining the AI models but we are encouraged by our early timing data and hope this will translate into an improved experience for our patients and those in lots of other cancer centres as well.”

Lead picture: Principal clinical scientist Rhydian Caines, fourth from left, with members of the brachytherapy and physics teams.

Read this report on the front page of the November 2024 issue of RAD Magazine.

You might also like

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read more