Communications of the ACM
Machine Learning to Help Physicians
Fraunhofer Institute researchers have developed software to help doctors interpret medical images to determine appropriate cancer treatment.
Credit: Fraunhofer MEVIS
New software from researchers at Germany's Fraunhofer Institute can aid physicians who must visually judge medical images to determine the course of cancer treatment.
The researchers say the program, which consists of modular processing components, is unique in that it uses deep learning to facilitate this task and reveal changes in images. Existing computer segmentation programs seek clearly defined image features, but this can often lead to errors that must be corrected by physicians, a process that can be time-consuming.
Fraunhofer researchers trained the software with computed tomography liver images from 149 patients and found the more data the program analyzed, the better it could automatically identify liver contours.
They say the new deep-learning approach should improve results and save physicians time.
"Our program package increases confidence during tumor measurement and follow-up," says Mark Schenk at the Fraunhofer Institute for Medical Image Computing. "The software can, for example, determine how the volume of a tumor changes over time and supports the detection of new tumors."
From Fraunhofer-Gesellschaft
View Full Article
Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA
No entries found