A team of researchers at the University of Michigan and the University of California, San Francisco in the United States developed a model supported by artificial intelligence that can determine in 10 seconds whether any part of a cancerous brain tumor that could be removed remains during surgery. The results of the new innovation were published in the journal Nature in November 13th.
The technology, called FastGlioma, outperformed traditional methods for identifying the remainder of the tumor by a large margin. When a neurosurgeon removes a life-threatening tumor from a patient’s brain, he is rarely able to remove the entire mass. Often, some cancerous cells are missed during the operation because surgeons are unable to distinguish between healthy parts of the brain and tumor remnants in the cavity where the mass was removed. The ability of residual tumor to resemble healthy brain tissue remains a major challenge in surgery.
Neurosurgical teams use different methods to locate residual tumor during the operation. They may need an MRI, which requires surgical machines that are not available everywhere. The surgeon may also use a fluorescent imaging agent to identify tumor tissue, which is not applicable to all types of tumors.
Catching remaining cancer cells
“FASTGlayoma is an artificial intelligence-based diagnostic system that has the potential to… “Transforming the field of neurosurgery by improving the comprehensive management of patients with metastatic brain tumors in the immediate future.”
In this international study of the technology, led by artificial intelligence, neurosurgical teams analyzed fresh, untreated samples taken from 220 patients who underwent surgery for diffuse glioma.
When compared to surgeries guided by FASTGlaoma prediction or image- and fluorescent-guided, the AI technology failed to detect high-risk residual tumor in only 3.8%, compared to an error rate of about 25% with traditional methods.
Artificial intelligence in the service of cancer patients
Rates of residual tumor after neurosurgery have not improved over the past 20 years. Residual tumor may not only lead to reduced quality of life and premature death for patients, but also increase the burden on a health system that expects to perform 45 million annual surgical procedures worldwide by 2030.
Global cancer initiatives have recommended the incorporation of new technologies, including advanced imaging methods and artificial intelligence, into cancer surgery.
Not only is FASTGlioma an accessible and affordable tool for neurosurgical teams working on gliomas, researchers say it can also accurately detect residual tumor for many non-glioma diagnoses, including pediatric brain tumors, such as medulloblastoma and glioma. Endothelial, meningioma.