Overview of the CLEF 2011 medical image classification and retrieval tasks

Jayashree Kalpathy-Cramer, Henning Muller, Steven Bedrick, Ivan Eggel, Alba G Seco De Herrera, Theodora Tsikrika

Research output: Chapter in Book/Report/Conference proceedingConference contribution

17 Citations (Scopus)

Abstract

The eighth edition of the ImageCLEF medical retrieval task was organized in 2011. A subset of the open access collection of PubMed Central was used as the database in 2011. This database contains 231,000 images and is substantially larger than previously used collections. Additionally, there was a larger fraction of non-clinical images such as graphs and charts. As in 2010, we had -three subtasks: modality classification, image-based and case-based retrieval. A new, simple hierarchy for article figures was created. Our belief is that the use of the detected modality should help filter out non-relevant images, thereby improving precision. The goal of the image-based retrieval task was to retrieve an ordered set of images from the collection that best meet the information need specified as a textual statement and a set of sample images, while the goal of the case-based retrieval task was to return an ordered set of articles (rather than images) that best meet the information need provided as a description of a case. The number of registrations to the medical task increased to 55 research groups. However, groups submitting runs have remained stable at 17, with the number of submitted runs increasing to 207. Of these, 130 were image-based retrieval runs, 43 were case-based runs while the remaining 34 were modality classification runs. Combining textual and visual cues most often led to best results, but results fusion needs to be used with care.

Original languageEnglish (US)
Title of host publicationCLEF 2011 - Working Notes for CLEF 2011 Conference
PublisherCEUR-WS
Volume1177
StatePublished - 2011
Event2011 Cross Language Evaluation Forum Conference, CLEF 2011 - Amsterdam, Netherlands
Duration: Sep 19 2011Sep 22 2011

Other

Other2011 Cross Language Evaluation Forum Conference, CLEF 2011
CountryNetherlands
CityAmsterdam
Period9/19/119/22/11

Fingerprint

Image classification
Image retrieval
Fusion reactions

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Kalpathy-Cramer, J., Muller, H., Bedrick, S., Eggel, I., De Herrera, A. G. S., & Tsikrika, T. (2011). Overview of the CLEF 2011 medical image classification and retrieval tasks. In CLEF 2011 - Working Notes for CLEF 2011 Conference (Vol. 1177). CEUR-WS.

Overview of the CLEF 2011 medical image classification and retrieval tasks. / Kalpathy-Cramer, Jayashree; Muller, Henning; Bedrick, Steven; Eggel, Ivan; De Herrera, Alba G Seco; Tsikrika, Theodora.

CLEF 2011 - Working Notes for CLEF 2011 Conference. Vol. 1177 CEUR-WS, 2011.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kalpathy-Cramer, J, Muller, H, Bedrick, S, Eggel, I, De Herrera, AGS & Tsikrika, T 2011, Overview of the CLEF 2011 medical image classification and retrieval tasks. in CLEF 2011 - Working Notes for CLEF 2011 Conference. vol. 1177, CEUR-WS, 2011 Cross Language Evaluation Forum Conference, CLEF 2011, Amsterdam, Netherlands, 9/19/11.
Kalpathy-Cramer J, Muller H, Bedrick S, Eggel I, De Herrera AGS, Tsikrika T. Overview of the CLEF 2011 medical image classification and retrieval tasks. In CLEF 2011 - Working Notes for CLEF 2011 Conference. Vol. 1177. CEUR-WS. 2011
Kalpathy-Cramer, Jayashree ; Muller, Henning ; Bedrick, Steven ; Eggel, Ivan ; De Herrera, Alba G Seco ; Tsikrika, Theodora. / Overview of the CLEF 2011 medical image classification and retrieval tasks. CLEF 2011 - Working Notes for CLEF 2011 Conference. Vol. 1177 CEUR-WS, 2011.
@inproceedings{eb47a262121c41c3a3a2092a1aee15b0,
title = "Overview of the CLEF 2011 medical image classification and retrieval tasks",
abstract = "The eighth edition of the ImageCLEF medical retrieval task was organized in 2011. A subset of the open access collection of PubMed Central was used as the database in 2011. This database contains 231,000 images and is substantially larger than previously used collections. Additionally, there was a larger fraction of non-clinical images such as graphs and charts. As in 2010, we had -three subtasks: modality classification, image-based and case-based retrieval. A new, simple hierarchy for article figures was created. Our belief is that the use of the detected modality should help filter out non-relevant images, thereby improving precision. The goal of the image-based retrieval task was to retrieve an ordered set of images from the collection that best meet the information need specified as a textual statement and a set of sample images, while the goal of the case-based retrieval task was to return an ordered set of articles (rather than images) that best meet the information need provided as a description of a case. The number of registrations to the medical task increased to 55 research groups. However, groups submitting runs have remained stable at 17, with the number of submitted runs increasing to 207. Of these, 130 were image-based retrieval runs, 43 were case-based runs while the remaining 34 were modality classification runs. Combining textual and visual cues most often led to best results, but results fusion needs to be used with care.",
author = "Jayashree Kalpathy-Cramer and Henning Muller and Steven Bedrick and Ivan Eggel and {De Herrera}, {Alba G Seco} and Theodora Tsikrika",
year = "2011",
language = "English (US)",
volume = "1177",
booktitle = "CLEF 2011 - Working Notes for CLEF 2011 Conference",
publisher = "CEUR-WS",

}

TY - GEN

T1 - Overview of the CLEF 2011 medical image classification and retrieval tasks

AU - Kalpathy-Cramer, Jayashree

AU - Muller, Henning

AU - Bedrick, Steven

AU - Eggel, Ivan

AU - De Herrera, Alba G Seco

AU - Tsikrika, Theodora

PY - 2011

Y1 - 2011

N2 - The eighth edition of the ImageCLEF medical retrieval task was organized in 2011. A subset of the open access collection of PubMed Central was used as the database in 2011. This database contains 231,000 images and is substantially larger than previously used collections. Additionally, there was a larger fraction of non-clinical images such as graphs and charts. As in 2010, we had -three subtasks: modality classification, image-based and case-based retrieval. A new, simple hierarchy for article figures was created. Our belief is that the use of the detected modality should help filter out non-relevant images, thereby improving precision. The goal of the image-based retrieval task was to retrieve an ordered set of images from the collection that best meet the information need specified as a textual statement and a set of sample images, while the goal of the case-based retrieval task was to return an ordered set of articles (rather than images) that best meet the information need provided as a description of a case. The number of registrations to the medical task increased to 55 research groups. However, groups submitting runs have remained stable at 17, with the number of submitted runs increasing to 207. Of these, 130 were image-based retrieval runs, 43 were case-based runs while the remaining 34 were modality classification runs. Combining textual and visual cues most often led to best results, but results fusion needs to be used with care.

AB - The eighth edition of the ImageCLEF medical retrieval task was organized in 2011. A subset of the open access collection of PubMed Central was used as the database in 2011. This database contains 231,000 images and is substantially larger than previously used collections. Additionally, there was a larger fraction of non-clinical images such as graphs and charts. As in 2010, we had -three subtasks: modality classification, image-based and case-based retrieval. A new, simple hierarchy for article figures was created. Our belief is that the use of the detected modality should help filter out non-relevant images, thereby improving precision. The goal of the image-based retrieval task was to retrieve an ordered set of images from the collection that best meet the information need specified as a textual statement and a set of sample images, while the goal of the case-based retrieval task was to return an ordered set of articles (rather than images) that best meet the information need provided as a description of a case. The number of registrations to the medical task increased to 55 research groups. However, groups submitting runs have remained stable at 17, with the number of submitted runs increasing to 207. Of these, 130 were image-based retrieval runs, 43 were case-based runs while the remaining 34 were modality classification runs. Combining textual and visual cues most often led to best results, but results fusion needs to be used with care.

UR - http://www.scopus.com/inward/record.url?scp=84922032536&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84922032536&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84922032536

VL - 1177

BT - CLEF 2011 - Working Notes for CLEF 2011 Conference

PB - CEUR-WS

ER -