Researchers have created a new artificial intelligence (AI) model that can identify the spread of metastatic brain cancer by analyzing MRI scans. This innovative approach provides critical information about patients’ cancer without needing invasive surgical procedures.
Researchers have created a new artificial intelligence (AI) model that can identify the spread of metastatic brain cancer by analyzing MRI scans. This innovative approach provides critical information about patients’ cancer without needing invasive surgical procedures.
The initial study, led by Dr. Matthew Dankner and Dr. Reza Forghani from McGill University, in collaboration with an international team of healthcare professionals and researchers, showed that the AI model can accurately identify cancer cells in adjacent brain tissue with an 85-percent success rate.
To validate the model, the researchers used MRI scans from more than 130 patients who underwent surgery to remove brain metastases at The Neuro (Montreal Neurological Institute-Hospital). They confirmed the AI’s accuracy by comparing its findings to microscopic examinations conducted by doctors on the tumor tissue.
Brain metastases, which are the most prevalent type of brain cancer, arise when cancer cells travel from other parts of the body to the brain. These tumors can become particularly aggressive when invasive cancer cells infiltrate surrounding healthy brain tissue, complicating treatment efforts.
“Our earlier studies found that invasive brain metastases are associated with shorter survival rates and a higher chance of tumor recurrence. These findings highlight the significant potential of machine learning to enhance our understanding of cancer and improve treatment options,” remarked Dankner, an Internal Medicine Resident at McGill and a post-doctoral researcher at the Rosalind & Morris Goodman Cancer Institute.
AI identifies subtle signs of cancer
The AI model recognizes slight changes in the adjacent brain tissue that suggest cancer has spread, detecting patterns that traditional imaging techniques may miss due to their reliance on human evaluation. This model was created in Forghani’s lab during his work at the Research Institute of the McGill University Health Centre and the University of Florida College of Medicine.
Earlier this year, the researchers discovered medications that might effectively treat certain brain metastases. However, identifying which patients might benefit from this treatment requires knowing whether cancer has infiltrated surrounding tissue. Surgery is the conventional approach, but it is not suitable for all patients, especially if tumors are located in difficult-to-reach areas or if the surgery poses significant health risks.
“With further advancements, our AI model could be incorporated into clinical practice, enabling earlier and more precise detection of cancer spread within the brain,” stated Dr. Benjamin Rehany, a Radiology Resident at the University of Toronto and one of the main authors of the study.
Although the research is still in its infancy, the team aims to broaden their study with larger datasets and enhance the AI model for practical clinical applications.
The research received funding support from the Canadian Cancer Society, the Canadian Institutes of Health Research, the Brain Canada Foundation, Health Canada, Fonds de recherche du Québec — Santé, and the Fondation de l’Association des radiologistes du Québec.
