Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.

© Springer International Publishing Switzerland 2014. We present a robust and high-throughput computational method for cell segmentation using multiplexed immunohistopathology images. The major challenges in obtaining an accurate cell segmentation from tissue samples are due to (i) complex cell and tissue morphology, (ii) different sources of variability including non-homogeneous staining and microscope specific noise, and (iii) tissue quality. Here we present a fast method that uses cell shape and scale information via unsupervised machine learning to enhance and improve general purpose segmentation methods. The proposed method is well suited for tissue cytology because it captures the the morphological and shape heterogeneity in different cell populations.We discuss our segmentation framework for analysing approximately one hundred images of lung and colon cancer and we restrict our analysis to epithelial cells.

Original publication





Publication Date





315 - 338