A new study led by a professor at York University’s Lassonde School of Engineering has found that an artificial intelligence (AI) method can help clinicians better distinguish between brain tumors and radiation necrosis on advanced MRI scans. The research, published in the International Journal of Radiation Oncology, Biology, Physics, involved collaboration with imaging scientists, neuro-oncologists, and neuro-radiologists at Sunnybrook Health Sciences Centre. Data was collected from over 90 cancer patients whose cancers had spread to the brain.
The study addresses a challenge faced after targeted radiation treatment for brain tumors. While stereotactic radiosurgery (SRS) can be effective in controlling tumors, it is not always successful. In up to 30 percent of cases, SRS does not stop tumor growth. Even when successful, the treatment may cause healthy brain tissue near the tumor to die off—a condition known as brain radiation necrosis—which can have significant side effects.
Researchers developed a three-dimensional deep learning AI model using two advanced attention mechanisms to tell apart tumor progression from radiation necrosis with a specialized MRI technique called chemical exchange saturation transfer (CEST). The AI model achieved over 85 percent accuracy in distinguishing between these conditions. In comparison, standard MRI methods correctly diagnose these conditions about 60 percent of the time; advanced MRI techniques alone increase this rate to around 70 percent.
"Differentiating tumor progression and radiation necrosis is very important - one needs more anti-cancer therapies and may need to be aggressively treated with more radiation, sometimes surgery. The other may require observation, anti-inflammatory drugs, so getting this right is crucial for patients," said Sadeghi-Naini.
