16 January, Waterloo, Canada (ANI): A computational model has been created by University of Waterloo researchers to more precisely predict the growth of deadly brain tumours.
A type of brain cancer called glioblastoma multiforme (GBM) has a one-year survival rate. It is difficult to treat because of its incredibly dense core, rapid growth, and location in the brain. Clinicians can benefit from estimating the diffusivity and proliferation rate of these tumours, but it is challenging to quickly and precisely estimate this data for a single patient.
St. Michael’s Hospital in Toronto is working with researchers from the Universities of Waterloo and Toronto to examine MRI data from a number of GBM patients. To more accurately anticipate the progression of cancer, they are utilising machine learning to comprehensively evaluate a patient’s tumour.
Five anonymous patients with GBM had two sets of MRIs analysed by researchers. After comprehensive MRIs, the patients got another round of MRIs after waiting a while. The patients’ MRIs gave the researchers a rare chance to comprehend how GBM grows when left unchecked because, for undisclosed reasons, they elected not to receive any treatment or intervention during this time.
The MRI data were transformed into patient-specific parameter estimates using a deep learning model, which the researchers then used to create a prediction model for GBM growth. They used this method to validate the model using synthetic and actual tumours from patients whose true characteristics were known.
As the study’s lead researcher and a PhD candidate in Applied Mathematics, Cameron Meaney stated, “We would have loved to do this analysis on a huge data set. However, given the nature of the illness, that’s very challenging because there isn’t a long life expectancy, and people tend to start treatment. The chance to compare five untreated tumours was therefore extremely uncommon and beneficial.
The scientists’ next step is to enlarge the model to account for the impact of treatment on the tumours now that they have a solid understanding of how GBM develops untreated. The number of MRIs in the data set would then rise from a few to thousands.
Meaney notes that patient outcomes can be significantly impacted by access to MRI data and collaboration between mathematicians and doctors.
The future of healthcare, according to Meaney, is the incorporation of quantitative analysis.
