- Thomas Deselaers (RWTH Aachen University, Computer Science Department)
- Henning Müller (University Hospitals of Geneva)
- Thomas M. Deserno (RWTH Aachen University, Department of Medical Informatics)
Special Issue finished
This special issue is finished and currently (as of July 2008) it is in press.
The list of papers published in this SI is
- Editorial by Thomas Deselaers, Henning Müller, Thomas M. Deserno
Title : Automatic Medical Image Annotation in ImageCLEF 2007: Overview, Results, and Discussion
Authors : Thomas Deselaers, Henning Mueller, Thomas M. Deserno
Title : Discriminative Cue Integration for Medical Image Annotation
Authors : Tatiana Tommasi, Francesco Orabona, and Barbara Caputo
Title : Hierarchical Classification Using a Frequency--Based Weighting and Simple Visual Features
Authors : Xin Zhou, Adrien Depeursinge, Henning Mueller
Title : Effectiveness of Global Features for Automatic Medical Image Classification and Retrieval - the experiences of OHSU at ImageCLEFmed
Authors : Jayashree Kalpathy-Cramer, William Hersh
Title : Deformations, Patches, and Discriminative Models for Automatic Annotation of Medical Radiographs
Authors : Thomas Deselaers, Hermann Ney
Title : A Framework and Baseline Results for the CLEF Medical Automatic Annotation Task
Authors : Mark O. Gueld, Christian Thies, Benedikt Fischer, Thomas M. Deserno
Title : Grayscale Medical Image Annotation using Local Relational Features
Authors : Lokesh Setia, Alexandra Teynor, Alaa Halawani, Hans Burkhardt
Title : Improving Efficiency and Effectiveness of the Image Distortion Model
Authors : Michael Springmann, Andreas Dander, Heiko Schuldt
Title : A Preliminary study on the relevance criteria for medical images
Authors : Shahram Sedghi, Mark Sanderson, Paul Clough
Call for Papers:
Automatic annotation of medical images is an important task allowing medical practitioners to access image data using textual queries. As part of ImageCLEF 2007, an automatic medical image annotation benchmark evaluation was performed with great interest.
The medical annotation task was first executed in ImageCLEF in 2005 and 2006. In 2007, the task comprised 11,000 training and 1,000 test images which had to be annotated according to a multi-axial, mono-hierarchical code.
With this special issue, we wish to summarize the efforts made in order to run the evaluations and the methods that were applied to solve the task. We wish to explore the effectiveness of the methods and give an insight into the problems that are still not solved:
- Are current image analysis techniques appropriate for automatic annotation of medical images?
- Do we need new techniques to solve this important issue? What are the challenges when it comes to medical images?
This special issue of Pattern Recognition Letters is inviting contributions from the participants of the ImageCLEF automatic medical image annotation tasks of 2005, 2006, and 2007. It will focus on the results from 2007 and the issues surrounding it.
Potential topics include (but are not limited to) the following:
- analysis of the results of the evaluations
- methods that were successfully applied to the task
- theoretical and/or practical insights into the problems for the task of medical image annotation
applicability of the techniques in praxis
Pattern Recognition Letters aims at rapid publication of concise articles of a broad interest in pattern recognition. Pattern Recognition Letters is published by Elsevier. More information on this journal can be found at http://www.elsevier.com/locate/patrec.
Papers should be submitted through the manuscript management system at http://ees.elsevier.com/patrec/. Formatted according to the guidelines available from that site. Papers should be submitted to the online system, selecting the "IMAGECLEF 2007" special issue as article type.