A team of computer scientists and pathologists at Stanford have developed a model for training computers to analyze breast cancer microscopic images, and findings published today in Science Translational Medicine show the machine analyses were more accurate than those conducted by humans.
In the paper, researchers describe their model, called Computational Pathologist (C-Path), a machine-learning-based method for automatically analyzing images of cancerous tissues and predicting patient survival. Researchers trained C-Path using existing tissue samples taken from patients whose prognosis was known. The computers analyzed the images, measured various tumor structures and used those structures to predict patient survival.
A comparison of the results against the known data showed the machines adapted the models to better predict survival and gradually determined what features of the cancers matter most and those that matter less in predicting survival. According to our release:
C-Path, in fact, assesses 6,642 cellular factors. Once trained using one group of patients, C-Path was asked to evaluate tissues of cancer patients it had not checked before and the result was compared against known data. Ultimately, C-Path yielded results that were a statistically significant improvement over human-based evaluation.
What's more, the computers identified structural features in cancers that matter as much or more than those that pathologists have focused on traditionally. In fact, they discovered that the characteristics of the cancer cells and the surrounding cells, known as the stroma, were both important in predicting patient survival.
Researchers say these findings add weight to what many in the scientific community have been contending for some time: that cancer is an "ecosystem," and that clinically significant information can be obtained by careful analysis of the complete tumor microenvironment.
Previously: Dramatic increase in number of older cancer survivors expected, Why do some cancers cause more financial problems than others? and A look at how best to care for America's growing population of cancer survivors
