INDIANAPOLIS — A team of Indiana University School of Medicine researchers has developed a new way to diagnose an elusive, deadly group of brain cancers using artificial intelligence models in place of costly, time-consuming molecular testing.
The group’s findings, published in Oxford Academic's Neuro-Oncology, lay the groundwork for technology's further use in diagnosing and treating adult-type diffuse gliomas. These types of brain cancer randomly spread into healthy brain tissue, making them exceedingly difficult to diagnose.
Conclusive diagnosis of gliomas currently requires complicated molecular testing.
"Molecular profiling is expensive, takes two to three weeks and may not be available in community hospitals," said Spyridon Bakas, PhD, senior corresponding author of the study, an associate professor and director of the Division of Computational Pathology at the IU School of Medicine and a researcher within the IU Melvin and Bren Simon Comprehensive Cancer Center.
"We developed an accurate and generalizable AI tool to classify such brain tumors using only histology images, offering a faster and more cost-effective alternative — even in limited resource environments, where molecular analysis is out of reach," Bakas said.
The team's AI model training and validation incorporated patient datasets from the United States, Italy, France, Australia, Austria and India. It met the latest World Health Organization (WHO) standards for conclusive diagnoses, and neuropathologists also verified the results.
Expediting diagnosis of these aggressive cancers is key, as it allows healthcare providers to more quickly begin treatment. The most aggressive form of diffuse glioma, glioblastoma, is the most common adult brain tumor. With a high recurrence rate, the average glioblastoma survival rate is about 15 months.
"If adopted widely, our AI model could make brain tumor diagnosis faster, more affordable and accessible, especially in regions where molecular testing is limited," Bakas said.
Shubham Innani, the first author of the study, a senior research analyst in the School of Medicine's Division of Computational Pathology and member of the Simon Cancer Center, said the new approach can complement existing pathology workflows and help deliver timely, equitable cancer care.
Bhakti Baheti, PhD, who was an assistant professor at the IU School of Medicine at the time of the study and is now faculty at Emory University School of Medicine, joined Bakas as a senior author on the study.
"By leveraging information across multiple magnifications, mimicking a pathologist’s workflow, AI can capture both microscopic details and broader context for more precise performance," Baheti said.
Neuropathologists W. Robert Bell, of the IU School of Medicine, and MacLean P. Nasrallah, of the University of Pennsylvania, rounded out the team.
This research was supported by funding from the National Cancer Institute of the National Institutes of Health and Lilly Endowment Inc. through its support for the IU Pervasive Technology Institute.
About the Indiana University School of Medicine
The IU School of Medicine is the largest medical school in the U.S. and is annually ranked among the top medical schools in the nation by U.S. News & World Report. The school offers high-quality medical education, access to leading medical research and rich campus life in nine Indiana cities, including rural and urban locations consistently recognized for livability. According to the Blue Ridge Institute for Medical Research, the IU School of Medicine ranks No. 13 in 2024 National Institutes of Health funding among all public medical schools in the country.
Writer: Rory Appleton, rapplet@iu.edu
For more news, visit the IU School of Medicine Newsroom: medicine.iu.edu/news