Robotic Eyes Spot Cancer

New software developed by researchers at Stanford University shows computers may be better than humans at diagnosing cancer, reports Popular Science's Rebecca Boyle. The system, called C-Path for Computational Pathologist, automatically evaluates microscopic images of breast cancer to determine its type and make a prognosis, Boyle says. And as the researchers found in their new study published in Science Translational Medicine, the algorithm is more accurate than a human doctor could be. C-Path "can classify the types of cancer cells present, and even identified a new set of features that are associated with a poor chance of survival," Boyle says. The researchers developed C-Path using existing tissue samples, and then human pathologists taught it to distinguish stromal and epithelial cells. When the researchers checked the machine's accuracy against a set of validation samples, "its results were a statistically significant improvement over human-based examination," Boyle adds. "C-Path even figured out something pathologists haven't — that the characteristics of the cancer cells and the surrounding cells were both important in determining a patient's outcome."