Fabian Rathke

Heidelberg Collaboratory for Image Processing (HCI), Room: H2.18
Universität Heidelberg
Speyerer Str. 6
D-69115 Heidelberg, Germany
Tel: +49 6221 54-8787
Email: name [dot] surename [at] iwr [dot] uni-heidelberg [dot] de

Research Projects

I am a PostDoc in the working group of Prof. Schnörr. My research interests are probabilistic graphical models, machine learning techniques, medical image analysis and non-parametric density estimation. ([CV] [Lebenslauf])

OCT retina segmentation

This project dealt with the segmentation of healthy and pathological retinal OCT scans. To this end we employed probabilistic graphical models, that could allowed us not only to infer a segmentation of the scan at hand, but also to value the quality of the segmentation as well as to estimate the degree of pathology.

The Matlab code of the project can be found here.

Non-parametric density estimation

Non-parametric density estimation with log-concave shape constraint is comparatively new research topic that has witnessed a lot attention in the last years. Here one wants to estimate a density function, whose logarithm is a concave function by maximizing the log-likelihood of the given data.

We proposed an estimation framework, which attacks the problem from a different perspective than the current state of the art approach. Its distinctive characteristic is a runtime which is largely independent on the size of the data set, allowing the estimation of densities for 10.000 or more data points (up to a million in 2-D).

A second journal paper and a corresponding software package for Matlab and R will be published soon.

Publications

F. Rathke and C. Schnörr. A Computational Approach to Log-Concave Density Estimation. An. St. Univ. Ovidius Constanta, 23(3):151–166, 2015 [pdf]
F. Rathke. Probabilistic Graphical Models for Medical Image Segmentation PhD thesis, University Heidelberg, 2015 [link] [pdf]
F. Rathke, S. Schmidt and C. Schnörr. Probabilistic Intra-Retinal Layer Segmentation in 3-D OCT Images Using Global Shape Regularization. Med. Image Anal., 18(5):781–794, 2014 [doi] [pdf]
F. Rathke, S. Schmidt and C. Schnörr. Order Preserving and Shape Prior Constrained Intra-Retinal Layer Segmentation. MICCAI 2011, Springer, vol. 6893, pp. 370–377, 2011 [doi] [pdf]
F. Rathke, K. Hansen, U. Brefeld and K.-R. Müller. StructRank: A New Approach for Ligand-Based Virtual Screening. J. Chem. Inf. Model., 51 (1), 83-92, 2011 [doi] [pdf]
K. Hansen, F. Rathke, T. Schroeter, G. Rast, T. Fox, J. M. Kriegl and S. Mika. Bias-Correction of Regression Models. A Case Study on hERG Inhibition. J. Chem. Inf. Model. , 49 (6), 1486-1496, 2009 [doi]
AttachmentSize
PDF icon main_elsevier.pdf5.09 MB
PDF icon JournalPopa.pdf700.56 KB
PDF icon miccai.pdf1.12 MB
PDF icon structrank.pdf494.35 KB
PDF icon Vitae.pdf69.58 KB
PDF icon Lebenslauf.pdf70.74 KB
PDF icon Thesis_Final_Print.pdf7.36 MB