Input. Just before Christmas, 24 projects received research funding from the Research Council of the Sahlgrenska Comprehensive Cancer Centre. One such project is led by Jonas Wallström, who is developing artificial intelligence models that could support future screening programs for prostate cancer.
The purpose of the grants distributed through the Sahlgrenska Comprehensive Cancer Center Research Council is to enhance the competitiveness of cancer research performed at Sahlgrenska University Hospital and Sahlgrenska Academy within the framework of the Strategic Research Plan. The grant budget for 2023 is approximately 5 million Swedish krona, which is distributed among the different projects. Here you can see a list of all 24 projects that received funding: https://www.gu.se/nyheter/24-projekt-far-stod-via-sccc
One of the researchers receiving funding is Jonas Wallström. He is a senior physician in abdominal and vascular radiology who conducts research in the areas of radiology and urology in the Department of Clinical Sciences, and has received research funding to develop artificial intelligence models associated with tumor progression.
How can artificial intelligence be used in cancer care?
– In this particular research project, we want to look at artificial intelligence models to evaluate the progression of tumor changes over time in men who have undergone repeat MRI scans. In the first step, we want to allow the model to sort out investigations without making serious findings. In a later step, we want to allow the model to try to predict image changes that could lead to the development of a dangerous cancer later on.
How can this method be used in clinical work?
-I have worked in the early detection of prostate cancer in the Gutenberg 2 screening study since the beginning in 2015 as a researcher and as a radiologist with evaluation of MRI images. We have shown that a screening algorithm with PSA testing followed by MRI reduces overdiagnosis of low-risk cancer by half while maintaining good detection of clinically important cancer, says Jonas Wallström, and continues:
The main challenge is how we will handle all images in a future screening program with limited access to expert reviewers, and here we believe that AI models can be part of the solution.
How will the research/project proceed in practice?
-We have a large database of MRI scans for training algorithms which can then be evaluated in an ongoing screening study.
Are there already similar examples?
– No, not really. We have a unique database of investigations in screening, and I don't know of any AI group working on follow-up, says Jonas Wallström.
How do you think this type of future care could benefit patients?
– By relieving and assisting the radiologist, a higher diagnostic quality can be provided to a greater number of patients, i.e. more equal care. I am involved in other AI-related research projects, but I do not currently use any algorithms in clinical work.
Written by: Sahlgrenskalev/Jacob Lundberg (Institute for Clinical Sciences)
Text was originally published on sahlgrenska.se
“Extreme tv maven. Beer fanatic. Friendly bacon fan. Communicator. Wannabe travel expert.”
More Stories
Key bacteria discovered – protect the intestinal mucus barrier in case of a low-fiber diet
The Saturday Chronicle: “Crazy. Or?” – Rebecca Widmo Ovill
Tobias from Oxlosund lives with obsessive thoughts: “There is a misconception about OCD”