taking notes of an experimental setup at ESRF
Dipl.-Inf. Univ. Lorenz König /køːniɡ̊/
| Natural languages: || New High German (first language),|
| Contact: || koenil01‹аτ›nyumc.org|
+1 (212) 26-33258
| Current location: || Center for Health Informatics and Bioinformatics|
NYU Medical Center
227 East 30th Street, 7th floor
New York, NY 10016
Collaborative Approaches to Image Analysis
Manual and semi-automatic segmentation still playing a considerable role in medical image analysis, the collaboration of more than one human observer (segmentor) may be desirable for several reasons, including split-up of workload, review of results by a more experienced observer, and estimation of multi-observer variability. To this end, a framework is currently being designed, allowing for management of original image data as well as multiple users’ contributions, and supporting consolidation into a concensus along with an estimate of its uncertainty. A prototype of a such framework is being implemented, including basic segmentation tools, thus creating the foundations for further refinement of collaborative approaches.
Analysis of Sub-Voxel Edge Localization Errors in Non-Ideal Modalities
Conventional sub-voxel edge detectors act on the assumption of a box-shaped voxel aperture
, or, using an equivalent formulation, point spread function
(PSF). This assumtion is quite well fulfilled by industrial digital optical cameras, thus not having been an issue in the application of sub-pixel edge detectors which in the beginning was focused on aerial imaging and materials testing.
However, modalities like MRI
have a significantly non-box voxel aperture, a fact that results in systematic edge localization errors (which we might call “edge aberration”). First results show that by an analysis and simple correction of these errors, edge localization can be improved by about half an order of magnitude.
see also: details below
Development of the “PaCaSe” Software
Originating from some of my student research projects (see below), the PaCaSe software
(“Patellar Cartilage Segmentation”) has been of great help to its users. It has hence been extended, now also comprising tools for thickness analysis and visualization and featuring input/output for a variety of data formats. PaCaSe will be superseded by the new collaborative framework (see above
see also: Semi-Automatic Patellar Cartilage Segmentation
||L. König, J. M. Raya Garcia del Olmo
Edge Aberration in MRI. Correction of Dislocations in Sub-Voxel Edge Detection—a Proof of Concept
Proc. of Bildverarbeitung für die Medizin (BVM 2011), Lübeck, Germany, March 2011.
The original publication is available online at www.springerlink.com.
||A. Horng, J. M. Raya Garcia del Olmo, M. Zscharn, L. König, M. Notohamiprodjo, M. Pietschmann, U. Hoehne-Hückstädt, I. Hermanns, U. Glitsch, R. Ellegast, K. G. Hering, M. Reiser, Ch. Glaser
Locoregional Deformation Pattern of the Patellar Cartilage After Different Loading Types—High-Resolution 3D-MRI Volumetry at 3 T in-vivo
Fortschr Röntgenstr. 2010; 182:1–9
||J. M. Raya Garcia del Olmo, A. Horng, L. König, M. Reiser, Ch. Glaser
Detecting Statistically Significant Changes in Cartilage Thickness with Sub-Voxel Precision
Proceedings of the 18th congress of the International Society for Magnetic Resonance in Medicine (ISMRM 2010), Stockholm, Sweden. Presentation 3192, electronic poster session “Meniscus & Cartilage”
||L. König, M. Groher, A. Keil, Ch. Glaser, M. Reiser, N. Navab
Semi-Automatic Segmentation of the Patellar Cartilage in MRI
Proc. of Bildverarbeitung für die Medizin (BVM 2007), Munich, Germany, March 2007. The original publication is available online at www.springerlink.com.
||S. Nissen-Meyer, L. König, M. Reiser
Maintaining Database Consistency in an Integrated, Heterogeneous HIS-RIS-PACS Environment
Proceedings of the 17th International Congress and Exhibition on Computer Assisted Radiology and Surgery (CARS 2003), London, UK, 25–28 June 2003
Student Research Projecs
Diplomarbeit “Digital Tomosynthesis in X-Ray Phase Contrast Imaging” at ESRF
on the applicability of the principle and notably the algorithms of digital tomosynthesis on the emerging imaging technique of x-ray phase contrast
advisors: Alberto Bravin, Paola Coan, Tobias Lasser
SEP “Assessment of Knee Cartilage Thickness using Magnetic Resonance Imaging”
advisor: Ben Glocker
IDP “Development and Implementation of a User-Friendly Tool for Analysis of the Patella Cartilage Using Magnetic Resonance Tomography”
advisor: Martin Groher et al.
Sub-voxel Edge Detection in Non-ideal Modalities
Detail of cylindrical phantom imaged in MRI slices perpendicular to phantom’s cylindrical axis. Six acquisitions, phantom moved to the right by 0.1 mm in-plane after each acquisition.
Fig. 1. Uncorrected Edges (click image to enlarge)
Fig. 2. Correction applied, otherwise same as fig. 1. (click image to enlarge)
Extracted edge of the phantom for each of the six different acquisitions, indicated by different colors.
Background: acquisition in which the phantom was in rightmost position.
Fig. 1. edge as identified by the original edge detector for each of the six acquisitions.
Fig. 2. same as fig. 1, but edge modified by heuristic correction.
More consistent shape of the edge in different positions qualitatively demonstrates improved edge precision after correction.