Colon Cancer Survival Prediction from Epithelium Nuclei Morphological Features Within Histology Slides Using Deep Learning

Gedeon, Rawan and Nagar, Atulya and Naguib, Raouf (2025) Colon Cancer Survival Prediction from Epithelium Nuclei Morphological Features Within Histology Slides Using Deep Learning. In: Practical Applications of Computational Biology and Bioinformatics, 18th International Conference (PACBB 2024). Lecture Notes in Networks and Systems, 1350 . Springer Nature, pp. 51-60. ISBN 9783031878725

[thumbnail of JIB_Paper_Last_Version_.pdf] Text
JIB_Paper_Last_Version_.pdf
Restricted to Repository staff only until 25 April 2026.

Download (12MB)

Abstract

Whole slide images (WSIs) are rich in prognostic information that can be quantified by means of deep learning and computer image analysis algorithms. In this study, we train a tissue classification network and a nuclei segmentation network. Then we use these models to process colon cancer patient cohort in order to evaluate the prognostic value of a risk index computed using morphological features for epithelium nuclei found in a tumour region to patient overall survival (OS). First, Convolutional Neural Networks (CNNs) were used to segment tissues in WSIs in a patch-wise approach. A tumour region is then localised in order to segment and classify the nuclei using U-Net segmentation network. Finally, we extract morphological features from the epithelium nuclei to predict patient OS. The generated risk index was shown to be significant to patient survival in a univariate and multivariate Cox analysis.

Item Type: Book Section
Additional Information and Comments: © 2025 The Author(s).
Faculty / Department: Faculty of Human and Digital Sciences > School of Computer Science and the Environment
SWORD Depositor: eprints api
Depositing User: eprints api
Date Deposited: 29 Jul 2025 08:10
Last Modified: 29 Jul 2025 08:10
URI: https://hira.hope.ac.uk/id/eprint/4705

Actions (login required)

View Item View Item