Master Thesis




In this project, we aspire to provide new Visual Analytics strategies that will support the correlative exploration and analysis of available prostate cancer data in a large cohort of prostate cancer patients. Analyzing intra-tumor tissue characteristics through Visual Analytics has been addressed multiple times in literature [1-4]. However, the comparison of tumor tissue characteristics across patients, still remains unanswered.

[1] E. Mörth, K. Wagner-Larsen, E. Hodneland, C. Krakstad, I. S. Haldorsen, S. Bruckner, and N. N. Smit. RadEx: Integrated Visual Exploration of Multiparametric Studies for Radiomic Tumor Profiling. Computer Graphics Forum, 2020.

[2] R. G. Raidou, U. van der Heide, C. V. Dinh, G. Ghobadi, J. F. Kallehauge, M. Breeuwer, and A. Vilanova. Visual Analytics for the Exploration of Tumor Tissue Characterization. Eurographics Conference on Visualization (EuroVis) 2015, 34(3):11–20, 2015.

[3] Lex, Alexander, Marc Streit, H‐J. Schulz, Christian Partl, Dieter Schmalstieg, Peter J. Park, and Nils Gehlenborg. "StratomeX: visual analysis of large‐scale heterogeneous genomics data for cancer subtype characterization." In Computer graphics forum, vol. 31, no. 3pt3, pp. 1175-1184. Oxford, UK: Blackwell Publishing Ltd, 2012.

[4] E. Mörth, I. S. Haldorsen, S. Bruckner, and N. N. Smit, "ParaGlyder: Probe-driven Interactive Visual Analysis for Multiparametric Medical Imaging Data," in Proceedings of Computer Graphics International, 2020, p. 351–363.


In this project, we want to investigate visual analytics strategies that support the flexible exploration and analysis of prostate cancer cohort data, targeting the comparison of tumor tissue characteristics across patients. We will follow a user-center strategy, which will target the integration of domain knowledge into an automated (unsupervised machine learning based) analytical approach to support cohort stratification of prostate cancer patients. Then, visual abstraction approaches will be required to facilitate the comparison of tumor profiles across patients.


  • Interest and knowledge in biomedical visualization and in visual analytics.
  • Knowledge and experience with (unsupervised) machine learning.
  • Good programming skills.
  • Creativity and enthusiasm.


To be discussed (depending on the background of the student), but a web-based environment is preferred.


For more information please contact Renata Raidou.