AI System Detects Genetic Mutations in Cancerous Brain Tumors
A new AI system is able to identify genetic mutations in cancerous brain tumors in under 90 seconds, potentially leading to improved diagnosis and treatment of gliomas.
Source: Getty Images
By - New research published in Nature Medicine showed that an artificial intelligence (AI)-based diagnostic screening system known as DeepGlioma can use rapid imaging to detect genetic mutations in cancerous brain tumors, possibly leading to enhanced diagnoses and treatment.
A glioma is cell growth that occurs in the brain or spinal cord, which can develop into cancerous or non-cancerous tumors, according to the Mayo Clinic. A study published in CA: A Cancer Journal for Clinicians shows that about 100,000 people worldwide receive a diffuse glioma diagnosis annually.
To help diagnose and treat gliomas, a group of neurosurgeons and engineers from Michigan Medicine have created an AI-based diagnostic screening system that uses rapid imaging to analyze tumor specimens to detect genetic mutations. Known as DeepGlioma, the system was developed in collaboration with investigators from New York University, the University of California, San Francisco, and other institutions.
Researchers noted that in the process of diagnosing and treating gliomas, molecular classification is critical. This is largely because the benefits and risks of surgery are inconsistent among brain tumor patients with variations in genetic makeup.
Previously, surgeons did not have access to a tool that differentiated diffuse gliomas during surgery. However, DeepGlioma can image brain tumor tissue through the combination of deep neural networks with an optical imaging method called stimulated Raman histology, according to the press release.
Following a study that included over 150 patients with diffuse glioma, researchers found that the system could define molecular subgroups of this condition with an accuracy exceeding 90 percent.
“DeepGlioma creates an avenue for accurate and more timely identification that would give providers a better chance to define treatments and predict patient prognosis,” said Todd Hollon, MD, lead study author and creator of DeepGlioma, a neurosurgeon at the University of Michigan Health and assistant professor of neurosurgery at U-M Medical School, in a press release.
Researchers also noted that since patients with malignant diffuse gliomas have a median survival time of only about 18 months, the rapid pace of a system like DeepGlioma is essential to enhancing treatment.
“Progress in the treatment of the most deadly brain tumors has been limited in the past decades- in part because it has been hard to identify the patients who would benefit most from targeted therapies,” said senior study author Daniel Orringer, MD, an associate professor of neurosurgery and pathology at NYU Grossman School of Medicine, who developed stimulated Raman histology, in a press release. “Rapid methods for molecular classification hold great promise for rethinking clinical trial design and bringing new therapies to patients.”
The use of AI for enhancing the detection of medical conditions is growing.
In March, researchers from several institutions created an AI-based app that could capture photos of skin lesions and provide insight into whether they derived from mpox.
Known as PoxApp, the AI was trained with a dataset of 130,000 images of various skin conditions. Through the app, patients can capture photos of abnormal skin lesions, get answers to questions, and receive a risk score along with recommendations for further action.
