Summary: Researchers have developed an AI model that detects metastatic brain cancer spread with 85% accuracy using MRI scans, offering a non-invasive alternative to surgery and potential advancements in cancer treatment.

Key Takeaways

  • AI detects cancer spread with high accuracy: The AI model identifies metastatic brain cancer spread in surrounding brain tissue with 85% accuracy using MRI scans, offering a non-invasive alternative to traditional methods.
  • Validated on real patient data: The model was tested on MRI scans from over 130 patients and validated against microscopic analysis of tumor tissue, demonstrating its reliability.
  • Potential for clinical impact: With further development, the AI could help detect cancer spread earlier and more accurately, aiding treatment decisions, especially for patients unable to undergo surgery.

—————————————————————————————————————————————————————————

Researchers have developed an artificial intelligence (AI) model that detects the spread of metastatic brain cancer using MRI scans, offering insights into cancer progression without invasive surgery.

In a proof-of-concept study, McGill University researchers Matthew Dankner, MD, PhD, and Reza Forghani, MD, PhD, along with an international team, demonstrated that the AI model detects cancer cells in surrounding brain tissue with 85% accuracy.

The model was tested on MRI scans from over 130 patients who underwent surgery to remove brain metastases at The Neuro (Montreal Neurological Institute-Hospital). Its accuracy was validated by comparing AI findings to microscopic tumor tissue analysis. Brain metastases, the most common form of brain cancer, occur when cancer spreads from other parts of the body. These aggressive tumors become harder to treat when cancer cells invade healthy brain tissue.

“Our previous research found that invasive brain metastases are linked to shorter survival and a higher risk of tumor regrowth,” says Dankner, an Internal Medicine Resident at McGill and post-doctoral researcher at the Rosalind & Morris Goodman Cancer Institute. “These findings demonstrate the enormous potential of machine learning to soon improve our understanding of cancer and its treatment.”

AI Detects Subtle Cancer Clues

The AI model identifies subtle changes in brain tissue that suggest cancer spread—patterns often too faint for traditional imaging methods. It was developed in Forghani’s lab during his time at the Research Institute of the McGill University Health Centre and the University of Florida College of Medicine.

Earlier this year, the team identified drugs that might treat some brain metastases, but doctors need to know if the cancer has spread to surrounding tissue to tailor treatments. Surgery is often required to assess this spread but isn’t always feasible for patients with inaccessible tumors or those in poor health.

“With further development, our AI model could become part of clinical practice, helping us detect cancer spread within the brain earlier and more accurately,” says Benjamin Rehany, MD, a radiology resident at the University of Toronto and one of the study’s primary authors.

While still in its early stages, the researchers aim to refine the AI model and expand their study using larger datasets to enable future clinical applications.