A number of papers have established that a high density of tumor-infiltrating lymphocytes (TILs) is highly correlated with a better prognosis for many different cancer types. More recently, some studies have shown that the spatial interplay between different subtypes of TILs (e.g. CD3, CD4, CD8) is more prognostic of disease outcome compared to just metrics related to TIL density. A challenge with TIL subtyping is that it relies on quantitative immunofluoresence or immunohistochemistry, complex and tissue-destructive technologies. In this paper we present a new approach called PhenoTIL to identify TIL sub-populations and quantify the interplay between these sub-populations and show the association of these interplay features with recurrence in early stage lung cancer. The approach comprises a Dirichlet Process Gaussian Mixture Model that clusters lymphocytes on H&E images. The approach was evaluated on a cohort of N=178 early stage non-small cell lung cancer patients, N=100 being used for model training and N=78 being used for independent validation. A Linear Discriminant Analysis classifier was trained in conjunction with 186 PhenoTIL features to predict the likelihood of recurrence in the test set. The PhenoTIL features yielded an AUC=0.84 compared to an approach involving just TIL density alone (AUC=0.58). In addition, a Kaplan-Meier analysis showed that the PhenoTIL features were able to statistically significantly distinguish early from late recurrence (p = 4 ∗ 10 −5 ).
Automatic detection and quantification of glands in gastric cancer may contribute to objectively measure the lesion severity, to develop strategies for early diagnosis, and most importantly to improve the patient categorization. This article presents an entire framework for automatic detection of glands in gastric cancer images. This approach starts by selecting gland candidates from a binarized version of the hematoxylin channel. Next, the gland’s shape and nuclei are characterized using local features which feed a Monte Carlo Cross validation method classifier trained previously with manually labeled images. Validation was carried out using a dataset with 1330 annotated structures (2372 glands) from seven fields of view extracted from gastric cancer whole slide images. Results showed an accuracy of 93% using a simple linear classifier. The presented strategy is quite simple, flexible and easily adapted to an actual pathology laboratory.