Researchers at Boston-based Brigham and Women's Hospital have developed a deep learning algorithm that can identify and outline a non-small cell lung cancer tumor on a CT scan within seconds, Science Daily reported Aug. 24.
Researchers used CT images from 787 patients to train their AI program to identify tumors and then asked eight radiation oncologists to edit outlines from the algorithm or another expert physician.
There was no significant difference between outlines drawn by physicians or the AI; however,when editing outlines created by the AI, physicians worked 65 percent faster and with 32 percent less variation.
They also rated the AI-drawn outlines more highly than the human ones.
"The benefits of this approach for patients include greater consistency in segmenting tumors and accelerated times to treatment. The clinician benefits include a reduction in mundane but difficult computer work, which can reduce burnout and increase the time they can spend with patients," corresponding author Raymond Mak, MD, of Brigham's Department of Radiation Oncology, said.