United States: A new AI program called FastGlioma can help doctors find and remove brain cancer that might be missed during surgery. It can calculate how much cancer is left after surgery with 92% accuracy, according to a study published on Nov. 13.
Fifty-seven percent of these were seen by both FastGlioma and the human doctor whereas the FastGlioma failed to identify high-risk residual tumor just beneath 4%, half-pie of the almost 25% miss rate seen in MRI scans or fluorescent dye detection by a human doctor.
As reported by HealthDay, not only that, but the AI can return these results within 10 seconds, which makes the program a potentially useful tool to surgeons when they are in the middle of removing a brain tumor, scientists reported. FastGlioma is an AI diagnostic model that can revolutionize neurosurgery by enhancing complete patient management of diffuse gliomas straight away, according to the study’s principal contributor Dr. Todd Hollon, a neurosurgeon at University of Michigan Health powerful aid to surgeons in the middle of removing a brain tumor, researchers said.
“FastGlioma is an artificial intelligence-based diagnostic system that has the potential to change the field of neurosurgery by immediately improving comprehensive management of patients with diffuse gliomas,” said senior researcher Dr. Todd Hollon, a neurosurgeon at University of Michigan Health.
“The technology itself is faster and more accurate than the state-of-art methods for tumor detection and it could be extended to other pediatric and adult brain tumor diagnoses,” said Dr Hollon in an interview with the university. “It might further be used as a reference model for direction of brain tumor surgery.”
In fact, neurosurgeons are seldom capable of resecting the whole big-size mass of life-threatening primary brain tumor; therefore, they leave back what doctors refer to residual tumor, the researchers pointed out in background notes.
This is leftover brain cancer because besides having the same appearance of healthy brain tissue along the margins of the cavity that results from the removal of a tumor, researchers said.
Of the two, Residual tumor causes cancer to recur to a person, reduces their lifespan and commonly warrants for further surgeries on the person’s brain, according to the researchers.
Even during a surgery, doctors attempt to identifyee residual tumor using MRI scans and fluorescent agents that light up tumor cells, but these have some value, researcher pointed out.
By integrating artificial intelligence into microscopic optical imaging to form FastGlioma, researchers enabled surgeons to remove a brain tumor more effectively.
The team prepared FastGlioma’s AI model from over 11 000 such surgeries and from 4 million separate microphotographs of healthy brain tissue and glioma tumors.
Also, the researchers than analyzed fresh unprocessed specimens’ samples from 220 patients who had operations of the brain cancers.
This Fast Glioma detected the residual tumor with up to 92 percent accuracy within about 100 seconds using the high-resolution images and with 90 percent of the accuracy when using a fast mode that relies on slightly lower resolution images.