Advertisement

New imaging technique may speed diagnosis of brain tumors

Stimulated Raman histologic images of diffuse astrocytoma (left) and meningioma (right), produced using a new imaging technique that may better diagnose brain tumores. Photo courtesy of Daniel Orringer/NYU Langone Health
Stimulated Raman histologic images of diffuse astrocytoma (left) and meningioma (right), produced using a new imaging technique that may better diagnose brain tumores. Photo courtesy of Daniel Orringer/NYU Langone Health

Jan. 6 (UPI) -- Researchers are using artificial intelligence -- combined with advanced optical imaging -- to rapidly diagnose brain tumors.

In a paper published Monday in the journal Nature Medicine, researchers described an imaging technique called stimulated Raman histology, or SRH, that reveals tumor infiltration in human tissue by collecting scattered laser light to illuminate features not typically visible in standard histologic images.

Advertisement

The new approach enables accurate, real-time intra-operative diagnosis of brain tumors, they wrote.

RELATED Brain tumor patient takes too much medical pot, lands in ER for observation

"With this imaging technology, cancer operations are safer and more effective than ever before," co-author Daniel A. Orringer, associate professor of neurosurgery at NYU Grossman School of Medicine, said in a press release. Orringer helped develop SRH and co-led the study with colleagues at the University of Michigan.

The technology works in concert with intra-operative MRI and fluorescence-guided surgery to provide high-resolution precision guidance for neurosurgeons. Microscopic images are processed and analyzed with artificial intelligence, and in less than two and a half minutes, surgeons are able to see a predicted brain tumor diagnosis.

Using the same technology, after the resection, they are able to accurately detect and remove otherwise undetectable tumor.

RELATED New brain cancer treatments showing promise in trials

"SRH will revolutionize the field of neuropathology by improving decision-making during surgery and providing expert-level assessment in the hospitals where trained neuropathologists are not available," said Matija Snuderl, an associate professor of pathology at NYU Grossman School of Medicine.

Advertisement

The Nature Medicine study assessed the diagnostic accuracy of brain tumor image classification using SRH against that of a pathologist's interpretation of conventional imaging. To build the AI tool, the authors trained a deep convolutional neural network with more than 2.5 million samples from 415 patients to classify tissue into 13 histologic categories that represent the most common brain tumors, including malignant glioma, lymphoma, metastatic tumors and meningioma.

In order to validate the accuracy of the tool, they enrolled 278 patients undergoing brain tumor resection or epilepsy surgery at three university medical centers. Brain tumor specimens were biopsied, split intra-operatively into sister specimens and randomly assigned to undergo standard diagnosis -- at a pathology lab, in a process that takes 20 to 30 minutes -- or diagnosis via the new approach.

RELATED Childhood brain tumor survivors face intellectual, financial struggles

The authors found that the results for both methods were comparable: the SRH-based diagnosis was 94.6 percent, compared with 93.9 percent for the pathologist-based interpretation. The authors believe the system's precise diagnostic capability could also be beneficial to centers that lack access to expert neuro-pathologists.

"As surgeons, we're limited to acting on what we can see," said Orringer. "This technology allows us to see what would otherwise be invisible, to improve speed and accuracy in the OR, and reduce the risk of misdiagnosis."

Advertisement

Latest Headlines