The approach, reported in the Proceedings of the National Academy of Sciences, is more accurate than the predictions of doctors who undergo years of highly-specialized training for the same purpose.
Gliomas are often fatal within two years of diagnosis, but some patients can survive for 10 years or more.
Therefore, predicting the course of a patient's disease at diagnosis is critical in selecting the right therapy and in helping patients and their families to plan their lives.
Doctors currently use a combination of genomic tests and microscopic examination of tissues to predict how a patient's disease will behave clinically or respond to therapy.
The reliable genomic testing cannot completely explain patient outcomes and microscopic examination is so subjective that different pathologists often providing different interpretations of the same case.
"There are large opportunities for more systematic and clinically meaningful data extraction using computational approaches," said Daniel J. Brat, the lead neuropathologist on the study, who began developing the software at the Winship Cancer Institute of Emory University.
The researchers used an approach called deep-learning to train the software to learn visual patterns associated with patient survival using microscopic images of brain tumor tissue samples.
When the software was trained using both images and genomic data, its predictions of how long patients survive beyond diagnosis were more accurate than those of human pathologists, according to researchers.
The researchers also demonstrated that the software learns to recognize many of the same structures and patterns in the tissues that pathologists use when performing their examinations.
The researchers are looking forward to future studies to evaluate whether the software can be used to improve outcomes for newly diagnosed patients. Xinhua