They concluded that the predicted contours were accurate enough to be implemented clinically with only minor or no changes.

In addition, when used on the Maverick supercomputer at the Texas Advanced Computer Center (TACC), the algorithm produced clinical target volumes in under a minute. TACC provided resources to shorten training time for the system and speed up research.


Cardenas says the aim of the project is to assist low- and middle-income countries which lack expertise in contouring, but that clinicians in the U.S. and other high-income nations could also utilize the approach to compare patient outcomes in different institutions and establish best practices for standards of care.

"I think it's going to change our field,” he said. “Some of these recommender systems are getting to be very good, and we're starting to see systems that can make predictions with a higher accuracy than some radiologists can. I hope that the clinical translation of these tools provides physicians with additional information that can lead to better patient treatments."

Research was carried out under the supervision of Laurence Court and with support from the National Institutes of Health.

The findings were published in a paper in the June 2018 edition of the International Journal of Radiation Oncology*Biology*Physics.

The head and neck are usually treated with different volumes for high, intermediate and low risk. The paper describes automating targets for high-risk areas. Additional ones will focus on low and intermediate predictions.

Neither Cardenas nor the other researchers involved responded for comment.

Back to HCB News

Rad Oncology Homepage