Researchers at Chalmers University of Technology in Sweden have carried out one of the largest studies to date using AI (artificial intelligence)-assisted image analysis of lymphoma.

“An AI-based computer system for interpreting medical images also contributes to increased equality in healthcare by giving patients access to the same expertise and being able to have their images reviewed within a reasonable time, regardless of which hospital they are in. Since an AI system has access to much more information, it also makes it easier in rare diseases where radiologists rarely see images,” says Ida Häggström, Associate Professor at the Department of Electrical Engineering at Chalmers.

Since an AI system has access to much more information, it also makes it easier in rare diseases where radiologists rarely see images.”

Together with clinically active researchers at, among others, Memorial Sloan Kettering Cancer Center in New York, Häggström has developed a computer model that was recently presented in The Lancet Digital Health. Based on more than 17,000 images from more than 5,000 lymphoma patients, the scientists have created a learning system in which computers have been trained to find visual signs of cancer in the lymphatic system.

The scientists have created a learning system in which computers have been trained to find visual signs of cancer in the lymphatic system.”

In the study, the researchers examined image archives that stretched back more than ten years. They compared the patients’ final diagnosis with scans from positron emission tomography (PET) and computed tomography (CT) taken before and after treatment. This information was then used to help train the AI computer model to detect signs of lymph node cancer in an image.

Deep learning system based on artificial intelligence

The computer model that Häggström has developed is called Lars, Lymphoma Artificial Reader System, and is a so-called deep learning system based on artificial intelligence. It works by inputting an image from positron emission tomography (PET) and analyzing this image using the AI model. It is trained to find patterns and features in the image, in order to make the best possible prediction of whether the image is positive or negative, i.e. whether it contains lymphoma or not.

 

A PET image is entered and analyzed by the AI model. Illustration: Ida Häggström/Chalmers

 

“I have used what is known as supervised training, where images are shown to the computer model, which then assesses whether the patient has lymphoma or not. The model also gets to see the true diagnosis, so if the assessment is wrong, the computer model is adjusted so that it gradually gets better and better at determining the diagnosis,” says Häggström.

Extensive clinical tests are needed

Häggström describes the process of teaching the computer to detect, in this case, cancer in the images as time-consuming, and says that it has taken several years to complete the study. One challenge has been to produce such a large amount of image material. It has also been challenging to adapt the computer model so that it can distinguish between cancer and the temporary treatment-specific changes that can be seen in the images after radiotherapy and chemotherapy.

“In the study, we estimated the accuracy of the computer model to be about ninety per cent, and especially in the case of images that are difficult to interpret, it could support radiologists in their assessments,” she says.

However, there is still a great deal of work to be done to validate the computer model if it is to be used in clinical practice.

“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,” says Häggström.

Source: Chalmers University of Technology in Sweden

Photo of Ida Häggström: Chalmers/Malin Arnesson