Paper Info
Title
The medGIFT Group in ImageCLEFmed 2011.
Abstract
This article presents the participation of the medGIFT group in ImageCLEFmed 2011. Since 2004, the group has participated in the medical image retrieval tasks of ImageCLEF each year. The main goal is to provide a baseline by using the same technology each year, and to search for further improvements in retrieval quality. There are three types of tasks for ImageCLEFmed 2011: modality classi- fication, image-based retrieval and case-based retrieval. The medGIFT group participated in all three tasks. For the image-based and case-based retrieval tasks, two existing retrieval engines were used: the GNU Image Finding Tool (GIFT) for visual retrieval and Apache Lucene for text. For the modality classification, a purely visual approach was used with GIFT for the visual retrieval and a kNN (k-Nearest Neighbors) classifier for the classification. Results show that the best text runs outperform the best visual runs by a factor of 10 in terms of mean average precision. Baselines provided by Apache Lucene and GIFT are ranked above the average among text runs and visual runs respectively in image-based retrieval. In the case- based retrieval task the Lucene baseline is the second best automatic run for text retrieval, and our mixed and visual runs are the best overall. For modality classification, GIFT and the kNN-based approach perform slightly better than the average of the visual approaches.
Year
Venue
DocType
2012
CLEF (Online Working Notes/Labs/Workshop)
Conference
Citations 
PageRank 
References 
0
0.34
0
Authors
4
Name
Order
Citations
PageRank
Dimitrios Markonis1516.52
Ivan Eggel220215.15
Alba García Seco de Herrera321616.48
Henning Müller49510.43