Reviewed by Lexie CornerMay 20 2024
Researchers at The Australian National University (ANU) have developed a new AI tool to classify brain tumors more quickly and accurately. The research has been published in Nature Medicine.
Dr. Danh-Tai Hoang. Image Credit: Dave Fanner/ The Australian National University
Dr. Danh-Tai Hoang asserts that accurate tumor diagnosis and classification are essential to successful patient care.
The current gold standard for identifying different kinds of brain tumors is DNA methylation-based profiling. DNA methylation acts like a switch to control gene activity, and which genes are turned on or off. But the time it takes to do this kind of testing can be a major drawback, often requiring several weeks or more when patients might be relying on quick decisions on therapies. There’s also a lack of availability of these tests in nearly all hospitals worldwide.
Dr. Danh-Tai Hoang, Study Author, The Australian National University
Researchers at ANU worked with specialists from the US National Cancer Institute to create DEPLOY, a method that predicts DNA methylation and divides brain tumors into ten main subtypes. DEPLOY uses histopathology images, which are minuscule photographs of a patient’s tissue.
The model was trained and verified using sizable datasets containing about 4,000 patients from the US and Europe.
Remarkably, DEPLOY achieved an unprecedented accuracy of 95 %. Furthermore, when given a subset of 309 particularly difficult to classify samples, DEPLOY was able to provide a diagnosis that was more clinically relevant than what was initially provided by pathologists. This shows the potential future role of DEPLOY as a complementary tool, adding to a pathologist’s initial diagnosis, or even prompting re-evaluation in the case of disparities.
Dr. Danh-Tai Hoang, Study Author, The Australian National University
The researchers anticipate that DEPLOY could eventually assist in classifying other types of cancer as well.
Journal Reference:
Hoang, D., et al. (2024) Prediction of DNA methylation-based tumor types from histopathology in central nervous system tumors with deep learning. Nature Medicine. doi.org/10.1038/s41591-024-02995-8.