Automated microscopic pathology can predict the prognosis of people with lung cancer, according to a study that applied machine learning methods to nearly 2,500 histopathology whole-slide images.

Investigators extracted close to 10,000 quantitative image features from the images and used regularized machine-learning methods to select the top features, and to distinguish shorter- and longer-term survivors with stage I adenocarcinoma or squamous cell carcinoma.

The automatically derived image features were able to predict prognosis, which the authors noted can contribute positively to precision oncology, thus improving outcomes. They added that the process can be used with images of other organs.

Citation: Yu K, Berry G, Altman R, Re C, Rubin D, Snyder M. Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features. [Published online ahead of print August 16, 2016]. Nat Commun. doi:10.1038/ncomms12474.