PT  - JOURNAL ARTICLE
AU  - Runyu Hong
AU  - Wenke Liu
AU  - David Fenyö
TI  - Predicting and Visualizing STK11 Mutation in Lung Adenocarcinoma Histopathology Slides Using Deep Learning
AID  - 10.1101/2020.02.20.956557
DP  - 2021 Jan 01
TA  - bioRxiv
PG  - 2020.02.20.956557
4099  - http://biorxiv.org/content/early/2021/12/15/2020.02.20.956557.short
4100  - http://biorxiv.org/content/early/2021/12/15/2020.02.20.956557.full
AB  - Studies have shown that STK11 mutation plays a critical role in affecting the lung adenocarcinoma (LUAD) tumor immune environment. By training an Inception-Resnet-v2 deep convolutional neural network model, we were able to classify STK11-mutated and wild type LUAD tumor histopathology images with a promising accuracy (per slide AUROC=0.795). Dimensional reduction of the activation maps before the output layer of the test set images revealed that fewer immune cells were accumulated around cancer cells in STK11-mutation cases. Our study demonstrated that deep convolutional network model can automatically identify STK11 mutations based on histopathology slides and confirmed that the immune cell density was the main feature used by the model to distinguish STK11-mutated cases.Competing Interest StatementThe authors have declared no competing interest.