Analysis of Spectral Images

Collaborators

Michael Hanselmann, Frederik Orlando Kaster, Claudia Beleites (TU Dresden), Nathalia Giese (Universitätsklinikum Heidelberg), Ron Heeren (AMOLF), Björn Menze (MIT CSAIL), Marc-André Weber (Universitätsklinikum Heidelberg), Peter Bachert (DKFZ), Oliver Nix (DKFZ), Patrik Zamecnik (DKFZ), Axel Walch (Helmholtz Zentrum München), Manfred Schmitt (Klinikum rechts der Isar), Fred A. Hamprecht.

• Institute for Atomic and Molecular Physics, Amsterdam
• Deutsches Krebsforschungszentrum, Heidelberg
• Technische Universität Dresden, Dresden
• Universitätsklinikum Heidelberg, Heidelberg
• MIT Computer Science and Artificial Intelligence Laboratory, Cambridge MA
• Helmholtz Zentrum München, München
• Klinikum rechts der Isar, München

Overview

Our research interest is the development of robust algorithms for the quantitative analysis of spectral images. We mainly focus on data aquired with mass spectrometry (MS), infrared spectroscopy (IR) and multimodal medical imaging. Depending on the application we work in the supervised, unsupervised or semi-supervised setting.

MS Imaging

The goal of mass spectrometry imaging (MSI) is to incorporate spatial context in the analysis of mass spectra. Challenges are the locating of homogeneous regions, the detection of underlying biological structures and the development of classification algorithms for cancer diagnosis and/or grading. In many cases the data is affected by a high noise level and only few label information is available.

IR imaging

The goal of this project is the classification of pancreatic cancer for grading on the basis of IR imaging data.

Multi-modal medical imaging

We aim to develop algorithms for automated cancer classification on multi modal image data, namely magnetic resonance spectroscopic imaging (MRSI), magnetic resonance imaging (MRI) and diffusion-weighted imaging (DWI). Main challenges are the fusion of multi resolution data and the lack of histological ground truth. As a practical application, the algorithms shall be used for evaluating the prognostic capabilities of different imaging modalities in primary brain tumour diagnostics.

Publications

• F.O. Kaster, B.M. Kelm, C.M. Zechmann, M.A. Weber, F.A. Hamprecht, O. Nix (2009). Classification of spectroscopic images in the DIROlab environment. Proc. of the World Congress 2009 in Medical Physics and Biomedical Engineering (11th International Congress of the IUPESM) (accepted for publication).
• M. Hanselmann, U. Köthe, B.Y. Renard, M. Kirchner, R.M.A. Heeren, F.A. Hamprecht (2009). Multivariate Watershed Segmentation of Compositional Data. Proc. of the 15th Int. Conf. on Discrete Geometry for Computer Imagery (DGCI) (accepted for publication).
• M. Hanselmann, U. Köthe, M. Kirchner, B.Y. Renard, E.R. Amstalden, K. Glunde, R.M.A. Heeren, F.A. Hamprecht (2009). Toward Digital Staining using Imaging Mass Spectrometry and Random Forests. Journal of Proteome Research 8 (7), 3558-3567
  [ pdf ]
• L. Görlitz, B. H. Menze, B. M. Kelm, F. A. Hamprecht (2009). Processing Spectral Images. Surface and Interface Analysis 41
  [ pdf ]
• B. H. Menze, B. M. Kelm, R. Masuch, U. Himmelreich, P. Bachert, W. Petrich, F. A. Hamprecht (2009). A Comparison of Random Forest and its Gini Importance with Standard Chemometric Methods for the Feature Selection and Classification of Spectral Data. BMC Bioinformatics, 10:213
  [ pdf ]
• B. M. Kelm, B. H. Menze, O. Nix, C. Zechmann, F. A. Hamprecht (2009). Estimating Kinetic Parameter Maps from Dynamic Contrast-Enhanced MRI using Spatial Prior Knowledge. IEEE Transaction on Medical Imaging
  [ pdf ]
• B. H. Menze, B. M. Kelm, M. Weber, P. Bachert, F. A. Hamprecht (2008). Mimicking the human expert: pattern recognition for an automated assessment of data quality in MRSI.Magnetic Resonance in Medicine 59, 1457-1466
  [ pdf ]
• M. Hanselmann, M. Kirchner*, B.Y. Renard*, E.R. Amstalden, K. Glunde, R.M.A. Heeren, F.A. Hamprecht (2008). Concise Representation of Mass Spectrometry Images. Analytical Chemistry 80(24), 9649-9658
  [ pdf ]
• B. H. Menze, W. Petrich, F. A. Hamprecht (2007). Multivariate feature selection and hierarchical classification for infrared spectroscopy: serum-based detection of bovine spongiform encephalopathy.Analytical and Bioanalytical Chemistry 387, 1801-1807
  [ pdf ]
• L. Görlitz, B. H. Menze, M.-A. Weber, B. M. Kelm, F. A. Hamprecht (2007). Semi-supervised tumor detection in magnetic resonance spectroscopic images using discriminative random fields. Proceedings of the 29th Symposium of the German Association for Pattern Recognition. Lecture Notes in Computer Science 4713, 224-233
• C. Zechmann, B. M. Kelm, P. Zamecnik, U. Ikinger, R. Waldherr, S. Röll, S. Delorme, F. A. Hamprecht, P. Bachert (2007). Can man still beat the machine? Automated vs. manual pattern recognition of 3D MRSI data of prostate cancer patients. Proceedings of the ISMRM 16th Scientific Meeting and Exhibition
  [ Preprint ]
• B. M. Kelm, B. H. Menze, C. M. Zechmann, K. T. Baudendistel, F. A. Hamprecht (2007). Automated Estimation of Tumor Probability in Prostate MRSI: Pattern Recognition vs. Quantification. Magnetic Resonance in Medicine 57, 150-159
  [ Preprint ]
• B. M. Kelm, N. Müller, B. H. Menze, F. A. Hamprecht (2006). Bayesian Estimation of Smooth Parameter Maps for Dynamic Contrast-Enhanced MR Images with Block-ICM. Proceedings of the Computer Vision and Pattern Recognition Workshop 2006 (Mathematical Methods in Biomedical Image Analysis), 96-103
  [ Preprint ]
• B. H. Menze, M. P. Lichy, P. Bachert, B. M. Kelm, H. P. Schlemmer, F. A. Hamprecht (2006). Optimal classification of long echo time in vivo magnetic resonance spectra in the detection of recurrent brain tumors. NMR in Biomedicine 19, 599-609
  [ Preprint ]
• B. M. Kelm, B. H. Menze, N. Müller, F. A. Hamprecht (2006). Estimation of pharmacokinetic parameters using spatial prior knowledge. Proceedings of the 23rd Annual Scientific Meeting of the ESMRMB No. 178, EPOS highlights session
• B. H. Menze, B. M. Kelm, F. A. Hamprecht (2006). Automated separation of low quality and artifact spectra by pattern recognition in the processing of MR spectral images. Proceedings of the ISMRM 14th Scientific Meeting and Exhibition
  [ Preprint ]
• B. M. Kelm, B. H. Menze, T. Neff, C. M. Zechmann, F. A. Hamprecht (2006). CLARET: a tool for fully automated evaluation of MRSI with pattern recognition methods. Bildverarbeitung für die Medizin 2006 - Algorithmen, Systeme, Anwendungen , 51-55 (Springer)
  [ Preprint ]
• B. H. Menze, B. M. Kelm, D. Heck, M. P. Lichy, F. A. Hamprecht (2006). Machine-based rejection of low quality spectra and estimation of brain tumor probabilities from magnetic resonance spectroscopic images. Bildverarbeitung für die Medizin 2006 - Algorithmen, Systeme, Anwendungen , 31-36 (Springer)
  [ Preprint ]
• M. P. Lichy, P. Bachert, F. A. Hamprecht, M.-A. Weber, J. Debus, D. Schulz-Ertner, H.-U. Kauczor, H.-P. Schlemmer (2006). Einsatz der 1H-MR-spektroskopischen Bildgebung in der Strahlentherapie: Cholin als Marker für die Bestimmung der relativen Wahrscheinlichkeit eines Tumorprogresses nach Bestrahlung glialer Hirntumoren. Zeitung für Röntgenforschung 178, 627-633
• B. M. Kelm, B. H. Menze, F. A. Hamprecht (2005). Automatische Lokalisation von Tumoren in 1H-NMR-spektroskopischen in vivo-Aufnahmen. GMA-Kongress 2005, VDI-Berichte 1883, 457-466
  [Preprint ]
• B. H. Menze, B. M. Kelm, M. P. Lichy, P. Bachert, H.-P. Schlemmer, F. A. Hamprecht (2005). Optimal processing in the automatic detection and localization of brain tumors using MRSI. Proceedings of the ISMRM 13th Scientific Meeting and Exhibition
• B. H. Menze, M. Wormit, P. Bachert, M. Lichy, H.-P. Schlemmer, F. A. Hamprecht (2004). Classification of in vivo magnetic resonance spectra. Classification, the ubiquitous challenge: Proceedings of GfKl 2004 (Springer)
• B. H. Menze, M. Wormit, P. Bachert, M. Lichy, H.-P. Schlemmer, F. A. Hamprecht (2004). Automated Classification of Magnetic Resonance Spectra for the Detection of Brain Tumors. 38. DGBMT Jahrestagung Biomedizinische Technik (BMT) , 438-439