Supporting the work usually done by CT and MRI-scans, a new program package from Fraunhofer Medical Institute will potentially increase confidence during tumor measurements and follow-ups.
The experts the Fraunhofer Institute for Medical Image Computing MEVIS in Bremen, Germany are expecting to demonstrate this new software in Chicago from 27th November – 2nd December at the world’s biggest radiology meeting, RSNA, Phys.org reports.
“Our program package increases confidence during tumor measurement and follow-up,” explains Mark Schenk from the Fraunhofer Institute said. “The software can, for example, determine how the volume of a tumor changes over time and supports the detection of new tumors”.
The machine applies deep learning that will reach further than what machine learning have done in the future, and this method is particularly helpful to conduct image segmentations done when experts designate exact organ outlines.
The already existing computer segmentation programs works to seek clearly defined image features such as certain gray values to detect a tumor. “However, this can often lead to errors,” according to Fraunhofer researcher Markus Harz. “The software assigns areas to the liver that do not belong to the organ.” These errors must be corrected by physicians, a process which can often be quite time-consuming.
“However, this can often lead to errors,” according to Fraunhofer researcher Markus Harz. “The software assigns areas to the liver that do not belong to the organ.” When these errors are detected assistance from a physician is often required, which is a time-consuming process. This new technology could potentially give the physicians more time doing other more important work, than correcting mistakes done by a machine.
The technology is self-learning and Fraunhofer explains how the scientists trained the software with CT liver images from 149 patients, and the result showed that the more the data analysed, the better it became.
Besides deep learning, the application will also use image registration, where software aligns images from different patient visits, to enable physicians to easily compare them. “Machine learning can aid the particularly difficult task of locating bone metastases in the torso in which hip bones, ribs, and spine are visible”, Phys. org writes.
“Currently, these metastases are often overlooked due to time constraints in clinical practice. Deep learning methods can help reliably discover metastases and thus improve treatment outcomes”.
Read more at: http://phys.org/news/2016-11-machine-physicians.html#jCp
Image Credit: Tumblr