In Finland, 900 brain tumors are diagnosed annually, in both children and adults. A brain tumor is a serious disease, but it can be treated, and the patient can survive. Researchers have been studying brain tumors for several decades, but there exist many challenges in the investigation and treatment of various types of tumors. In the future, artificial intelligence (AI) can enable more personalized and more efficient treatment of brain tumors. Adjunct Professor Joonas Haapasalo, M.D, Ph.D., is leading a study of brain tumors in Tampere University Hospital in Finland, wherein AI is used to identify various factors involved in the diagnosis and treatment.
In Finland, brain tumor research has been based on research in tumor biology for decades, and extensive knowledge has been accumulated in this regard. With advances in medicine and research, personalized assessment and treatment of patients is expected to become increasingly feasible. Thus, researchers are attempting to identify tailored treatment protocols for brain tumors more efficiently.
One of these researchers is Joonas Haapasalo whose research in Artificial Intelligence for brain tumor treatments can enable more personalized treatment for patients. For example, Haapasalo aims to improve brain tumor treatments by decreasing unwanted side effects and to improve the methods used in the diagnosis of tumors. Haapasalo is part of Professor Michael Taylor’s pediatric brain tumor research group in the Hospital for Sick Children in Toronto, Canada. At the same time, Haapasalo is leading the clinical brain tumor research at Tampere University Hospital in Finland. Haapasalo also serves as the coordinator of the Brain Tumor Pilot for National Neurocenter in Finland.
Artificial Intelligence advances personalized brain tumor research
The aim of the study is to create a comprehensive research database for patients with brain tumors using AI. The primary aim is to reduce the time-consuming work of researchers who have to go through the patient journals and data manually. In the future, artificial intelligence could generate such data, in other words, the algorithm would automatically generate patient information for brain tumor research.
The functionality of the artificial intelligence algorithm can be established by comparing the data generated by the algorithm with the data created by the researcher. If these two are alike, researchers can start using the algorithm generally in brain tumor research. In the best-case scenario, a nationwide brain tumor registry could be created with the help of AI.
The study is nearly complete for wider test use
An aim of the study was to collect results that can be linked to tumor biology and thus to develop personalized diagnostics and treatment. In the past, there have been many categories of brain tumors that have appear similar in microscopic evaluations; however, increased knowledge on genetics and gene regulation has taught us that these tumors have completely different biological backgrounds. As a result, tumor diagnostics have become more sophisticated in recent years, and these new methods have made it possible to develop more personalized treatments.
According to Dr. Haapasalo, with the AI algorithm, researchers could gather clinical variables from patients with brain tumors with better survival. Such data could be analyzed further to identify patterns and factors related to the individual’s characteristics that could explain the longer survival time, for example due to the patient’s more effective immune response.
– In addition, it is known that patients react differently to cancer treatments. If this could be evaluated in more detail, it would help to create more personalized therapeutic strategies.
The research needs to be interdisciplinary to ensure enough high-quality data. In addition to medicine, data scientists and legal professionals play an important role in the success of this pioneering AI research.
International frontiers
The research is carried out in cooperation with Finnish universities, university hospitals, and International hospitals. The international research partners include The Hospital for Sick Children in Toronto, Canada, Heidelberg University in Germany, and The University of Sydney in Australia.
“The research needs to be interdisciplinary to ensure enough high-quality data,” Haapasalo states. In addition to medicine, data scientists and legal professionals play an important role in the success of this pioneering AI research.
Image: MRI of the brain displaying a malignant brain tumor on the left side