AI technology detects cancerous brain tumors in 10 seconds during surgery
A groundbreaking artificial intelligence tool called FastGlioma has been developed that allows surgeons to detect residual cancerous brain tumors within 10 seconds during surgery. The innovation, described in a recent study in Nature, is seen as a significant advance in neurosurgery and outperforms traditional tumor detection methods. Researchers from the University of Michigan and the University of California, San Francisco led the study and highlighted its potential to improve surgical outcomes for patients with diffuse gliomas.
Todd Hollon, MD, a neurosurgeon at University of Michigan Health, described FastGlioma as a transformative diagnostic tool that provides a faster and more accurate method for identifying tumor remnants. He noted that it may reduce reliance on current methods, such as intraoperative MRI or fluorescent imaging, which are often inaccessible or inappropriate for all tumor types.
Addressing residual tumors during surgery
According to the study from Michigan Medicine – University of Michigan, residual tumors, which often resemble healthy brain tissue, are a common challenge in neurosurgery. Surgeons have traditionally struggled to distinguish between a healthy brain and the remaining cancerous tissue, leading to incomplete removal of the tumor. FastGlioma addresses this by combining high-resolution optical imaging with artificial intelligence to quickly and accurately identify tumor infiltration.
In an international study, the model was tested on samples from 220 patients with low- or high-grade diffuse gliomas. FastGlioma achieved an average accuracy of 92%, which significantly outperformed conventional methods, which had a higher error rate for high-risk tumor remnants. Co-senior author Shawn Hervey-Jumper, MD, professor of neurosurgery at UCSF, highlighted its ability to improve surgical precision while minimizing reliance on imaging agents or time-consuming procedures.
Future applications in cancer surgery
FastGlioma is based on basic models, a type of AI trained on huge data sets, allowing adaptation to different tasks. The model has shown potential for application in other cancers, including lung, prostate and breast tumors, without the need for extensive retraining.
Aditya S. Pandey, MD, chair of neurosurgery at the University of Michigan, affirmed his role in improving surgical outcomes worldwide, in line with recommendations to integrate AI into cancer surgery. Researchers aim to expand its use to other tumor types, potentially reshaping cancer treatment worldwide.
