Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features.

Yu KH, Zhang C, Berry GJ, Altman RB, Ré C, Rubin DL, Snyder M. · 2016

Read Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features. by Yu KH, Zhang C, Berry GJ, Altman RB, Ré C, Rubin DL, Snyder M. (2016) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc4990706.txt.

Description

Kun-Hsing Yu and six co-authors teach a computer to predict lung cancer survival from microscope images alone. Their study appeared in Nature Communications in 2016. Pathologists read tumor slides by eye to diagnose non-small cell lung cancer, but their calls predict survival poorly. The team gathered more than two thousand whole slide images from a cancer atlas, plus a Stanford tissue array set. Software extracted nearly ten thousand quantitative features describing cell shapes, textures, and tissue patterns.

Machine learning picked the features that split shorter term from longer term survivors in stage I disease. The split held for both adenocarcinoma and squamous cell carcinoma, the two main types. It also validated on the independent Stanford set. The authors argue automated image features can feed precision oncology, adding an objective prognosis layer to the standard slide reading. The method, they note, transfers to other organs.

Catalog record

CREATORYu KH, Zhang C, Berry GJ, Altman RB, Ré C, Rubin DL, Snyder M.
YEAR2016
COLLECTIONPAPERS
FORMATTXT
FILE SIZE51 KB
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC4990706

Browse the full library · About the Tower of Babel