Algorithm Reduces NSCLC Tumor Segmentation Times by 65%


A newly developed deep learning algorithm has the potential to transform a historically time-consuming process into a task that only takes seconds, saving valuable time for clinicians treating non-small lung cancer patients. cells.

The algorithm is said to accurately identify and segment tumors visualized on CT scans. A new analysis published in Lancet Digital Health showed that physicians using the algorithm were able to accurately target tissue for radiation therapy planning sometimes up to 65% faster than traditional manual methods.

Corresponding study author Raymond Mak, MD, of Brigham’s Department of Radiation Oncology, and his colleagues explained how this new AI method benefits patients and clinicians:

“The benefits of this approach for patients include greater consistency in tumor segmentation and accelerated treatment times. Benefits for clinicians include a reduction in mundane but difficult IT work, which can reduce burnout and increase the time they can spend with patients.

Working closely with radiation oncologists, the experts trained their model on CT lung images of 787 patients and tested on CT scans of more than 1,300 patients from external datasets. To examine the segmentation performance of the algorithm, eight radiation oncologists were asked to perform manual segmentations on images and then blindly evaluate the segmentations performed by other physicians and the algorithm.


Comments are closed.