Automatic image modality based classification and annotation to improve medical image retrieval

Jayashree Kalpathy-Cramer, William (Bill) Hersh

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

Abstract

Medical image retrieval can play an important role for diagnostic and teaching purposes in medicine. Image modality is an important visual characteristic that can be used to improve retrieval performance. Many test and online collections do not contain information about the image modality. We have created an automatic image classifier for both grey-scale and colour medical images. We evaluated the performance of the two modality classifiers, one for grey-scale images and the other for colour images on the CISMeF and the ImageCLEFmed 2006 databases. Both classifiers were created using a neural network architecture for learning. Low level colour and texture based feature vectors were extracted to train the network. Both classifiers achieved an accuracy of > 95% on the test collections that they were tested on. We also evaluated the performance of these classifiers on a selection of queries from the ImageCLEFmed 2006. The precision of the results was improved by using the modality classifier to resort the results of a textual query.

Original languageEnglish (US)
Title of host publicationStudies in Health Technology and Informatics
Pages1334-1338
Number of pages5
Volume129
StatePublished - 2007
Event12th World Congress on Medical Informatics, MEDINFO 2007 - Brisbane, QLD, Australia
Duration: Aug 20 2007Aug 24 2007

Other

Other12th World Congress on Medical Informatics, MEDINFO 2007
CountryAustralia
CityBrisbane, QLD
Period8/20/078/24/07

Fingerprint

Image retrieval
Classifiers
Color
Teaching
Medicine
Learning
Databases
Network architecture
Textures
Neural networks

Keywords

  • content-based image retrieval
  • image annotation
  • medical imaging
  • neural networks

ASJC Scopus subject areas

  • Biomedical Engineering
  • Health Informatics
  • Health Information Management

Cite this

Kalpathy-Cramer, J., & Hersh, W. B. (2007). Automatic image modality based classification and annotation to improve medical image retrieval. In Studies in Health Technology and Informatics (Vol. 129, pp. 1334-1338)

Automatic image modality based classification and annotation to improve medical image retrieval. / Kalpathy-Cramer, Jayashree; Hersh, William (Bill).

Studies in Health Technology and Informatics. Vol. 129 2007. p. 1334-1338.

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

Kalpathy-Cramer, J & Hersh, WB 2007, Automatic image modality based classification and annotation to improve medical image retrieval. in Studies in Health Technology and Informatics. vol. 129, pp. 1334-1338, 12th World Congress on Medical Informatics, MEDINFO 2007, Brisbane, QLD, Australia, 8/20/07.
Kalpathy-Cramer J, Hersh WB. Automatic image modality based classification and annotation to improve medical image retrieval. In Studies in Health Technology and Informatics. Vol. 129. 2007. p. 1334-1338
Kalpathy-Cramer, Jayashree ; Hersh, William (Bill). / Automatic image modality based classification and annotation to improve medical image retrieval. Studies in Health Technology and Informatics. Vol. 129 2007. pp. 1334-1338
@inproceedings{3a99cf64159f44c182c0e3fda89d42cd,
title = "Automatic image modality based classification and annotation to improve medical image retrieval",
abstract = "Medical image retrieval can play an important role for diagnostic and teaching purposes in medicine. Image modality is an important visual characteristic that can be used to improve retrieval performance. Many test and online collections do not contain information about the image modality. We have created an automatic image classifier for both grey-scale and colour medical images. We evaluated the performance of the two modality classifiers, one for grey-scale images and the other for colour images on the CISMeF and the ImageCLEFmed 2006 databases. Both classifiers were created using a neural network architecture for learning. Low level colour and texture based feature vectors were extracted to train the network. Both classifiers achieved an accuracy of > 95{\%} on the test collections that they were tested on. We also evaluated the performance of these classifiers on a selection of queries from the ImageCLEFmed 2006. The precision of the results was improved by using the modality classifier to resort the results of a textual query.",
keywords = "content-based image retrieval, image annotation, medical imaging, neural networks",
author = "Jayashree Kalpathy-Cramer and Hersh, {William (Bill)}",
year = "2007",
language = "English (US)",
isbn = "9781586037741",
volume = "129",
pages = "1334--1338",
booktitle = "Studies in Health Technology and Informatics",

}

TY - GEN

T1 - Automatic image modality based classification and annotation to improve medical image retrieval

AU - Kalpathy-Cramer, Jayashree

AU - Hersh, William (Bill)

PY - 2007

Y1 - 2007

N2 - Medical image retrieval can play an important role for diagnostic and teaching purposes in medicine. Image modality is an important visual characteristic that can be used to improve retrieval performance. Many test and online collections do not contain information about the image modality. We have created an automatic image classifier for both grey-scale and colour medical images. We evaluated the performance of the two modality classifiers, one for grey-scale images and the other for colour images on the CISMeF and the ImageCLEFmed 2006 databases. Both classifiers were created using a neural network architecture for learning. Low level colour and texture based feature vectors were extracted to train the network. Both classifiers achieved an accuracy of > 95% on the test collections that they were tested on. We also evaluated the performance of these classifiers on a selection of queries from the ImageCLEFmed 2006. The precision of the results was improved by using the modality classifier to resort the results of a textual query.

AB - Medical image retrieval can play an important role for diagnostic and teaching purposes in medicine. Image modality is an important visual characteristic that can be used to improve retrieval performance. Many test and online collections do not contain information about the image modality. We have created an automatic image classifier for both grey-scale and colour medical images. We evaluated the performance of the two modality classifiers, one for grey-scale images and the other for colour images on the CISMeF and the ImageCLEFmed 2006 databases. Both classifiers were created using a neural network architecture for learning. Low level colour and texture based feature vectors were extracted to train the network. Both classifiers achieved an accuracy of > 95% on the test collections that they were tested on. We also evaluated the performance of these classifiers on a selection of queries from the ImageCLEFmed 2006. The precision of the results was improved by using the modality classifier to resort the results of a textual query.

KW - content-based image retrieval

KW - image annotation

KW - medical imaging

KW - neural networks

UR - http://www.scopus.com/inward/record.url?scp=70549090598&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70549090598&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781586037741

VL - 129

SP - 1334

EP - 1338

BT - Studies in Health Technology and Informatics

ER -