Overview of the ImageCLEF 2013 medical tasks

In 2013, the tenth edition of the medical task of the Image-CLEF benchmark was organized. For the first time, the ImageCLEFmedworkshop takes place in the United States of America at the annualAMIA (American Medical Informatics Association) meeting even thoughthe task was organized as in previous years in connection with the otherImageCLEF tasks. Like 2012, a subset of the open access collection ofPubMed Central was distributed. This year, there were four subtasks:modality classification, compound figure separation, image–based andcase–based retrieval. The compound figure separation task was includeddue to the large number of multipanel images available in the literatureand the importance to separate them for targeted retrieval. More com-pound figures were also included in the modality classification task tomake it correspond to the distribution in the full database. The retrievaltasks remained in the same format as in previous years but a largernumber of tasks were available for image–based and case–based tasks.This paper presents an analysis of the techniques applied by the tengroups participating 2013 in ImageCLEFmed.