Swedish researchers develop new AI computer model to detect lymphatic cancer

Swedish researchers develop new AI computer model to detect lymphatic cancer Jen Brogan

In the study, the LARS system accurately detected 90% of lymphatic cancers

Researchers from Chalmers University of Technology in Sweden have developed a new computer model using artificial intelligence (AI), which successfully identifies signs of lymphatic cancer.

The model was developed in collaboration with researchers from Memorial Sloan Kettering Cancer Center, Chalmers University of Technology, Medical University in Vienna, Icahn School of Medicine at Mount Sinai and NYU Langone Health, with results published in The Lancet Digital Health.

Lymphoma is a cancer of the lymphatic system, including the lymph nodes, spleen, thymus gland and bone marrow, and can affect other organs throughout the body.

The two main subtypes of lymphoma are Hodgkin’s lymphoma and non-Hodgkin’s lymphoma, which is the sixth most common cancer in the UK, responsible for around 14,200 cases every year, according to Cancer Research UK.

Using AI-assisted image analysis of lymphoma, researchers developed a deep learning system to train computers based on over 17,000 images from more than 5,000 lymphoma patients to spot visual signs of cancer in the lymphatic system.

The Lymphoma Artificial Reader System (LARS) works by inputting an image from positron emission tomography and analysing the image using the model to identify patterns and features in the image to confirm whether it contains lymphoma or not.

Results showed that LARS accurately detected about 90% of lymphatic cancers, which could help support radiologists, particularly when analysing images that are difficult to interpret, in their assessments.

Ida Häggström, associate professor, department of electrical engineering, Chalmers University of Technology, said: “We have created a learning system in which computers have been trained to find visual signs of cancer in the lymphatic system” that could increase “equality in healthcare by giving patients access to the same expertise and being able to have their images reviewed within a reasonable time”.

She continued: “We have made the computer code available now so that other researchers can continue to work on the basis of our computer model, but the clinical tests that need to be done are extensive.”