Classification of diabetic retinopathy images using multi-class multiple-instance learning based on color correlogram features

Ragav Venkatesan, Parag Chandakkar, Baoxin Li, Helen K. Li

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

21 Citations (Scopus)

Abstract

All people with diabetes have the risk of developing diabetic retinopathy (DR), a vision-threatening complication. Early detection and timely treatment can reduce the occurrence of blindness due to DR. Computer-aided diagnosis has the potential benefit of improving the accuracy and speed in DR detection. This study is concerned with automatic classification of images with microaneurysm (MA) and neovascularization (NV), two important DR clinical findings. Together with normal images, this presents a 3-class classification problem. We propose a modified color auto-correlogram feature (AutoCC) with low dimensionality that is spectrally tuned towards DR images. Recognizing the fact that the images with or without MA or NV are generally different only in small, localized regions, we propose to employ a multi-class, multiple-instance learning framework for performing the classification task using the proposed feature. Extensive experiments including comparison with a few state-of-art image classification approaches have been performed and the results suggest that the proposed approach is promising as it outperforms other methods by a large margin.

Original languageEnglish (US)
Title of host publicationProceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
Pages1462-1465
Number of pages4
DOIs
StatePublished - 2012
Event34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2012 - San Diego, CA, United States
Duration: Aug 28 2012Sep 1 2012

Other

Other34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2012
CountryUnited States
CitySan Diego, CA
Period8/28/129/1/12

Fingerprint

Diabetic Retinopathy
Color
Learning
Computer aided diagnosis
Image classification
Medical problems
Blindness
Experiments
Microaneurysm

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Biomedical Engineering
  • Health Informatics

Cite this

Venkatesan, R., Chandakkar, P., Li, B., & Li, H. K. (2012). Classification of diabetic retinopathy images using multi-class multiple-instance learning based on color correlogram features. In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS (pp. 1462-1465). [6346216] https://doi.org/10.1109/EMBC.2012.6346216

Classification of diabetic retinopathy images using multi-class multiple-instance learning based on color correlogram features. / Venkatesan, Ragav; Chandakkar, Parag; Li, Baoxin; Li, Helen K.

Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. 2012. p. 1462-1465 6346216.

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

Venkatesan, R, Chandakkar, P, Li, B & Li, HK 2012, Classification of diabetic retinopathy images using multi-class multiple-instance learning based on color correlogram features. in Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS., 6346216, pp. 1462-1465, 34th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2012, San Diego, CA, United States, 8/28/12. https://doi.org/10.1109/EMBC.2012.6346216
Venkatesan R, Chandakkar P, Li B, Li HK. Classification of diabetic retinopathy images using multi-class multiple-instance learning based on color correlogram features. In Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. 2012. p. 1462-1465. 6346216 https://doi.org/10.1109/EMBC.2012.6346216
Venkatesan, Ragav ; Chandakkar, Parag ; Li, Baoxin ; Li, Helen K. / Classification of diabetic retinopathy images using multi-class multiple-instance learning based on color correlogram features. Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS. 2012. pp. 1462-1465
@inproceedings{f726c0c397794e63a6eae1ca6a58aac3,
title = "Classification of diabetic retinopathy images using multi-class multiple-instance learning based on color correlogram features",
abstract = "All people with diabetes have the risk of developing diabetic retinopathy (DR), a vision-threatening complication. Early detection and timely treatment can reduce the occurrence of blindness due to DR. Computer-aided diagnosis has the potential benefit of improving the accuracy and speed in DR detection. This study is concerned with automatic classification of images with microaneurysm (MA) and neovascularization (NV), two important DR clinical findings. Together with normal images, this presents a 3-class classification problem. We propose a modified color auto-correlogram feature (AutoCC) with low dimensionality that is spectrally tuned towards DR images. Recognizing the fact that the images with or without MA or NV are generally different only in small, localized regions, we propose to employ a multi-class, multiple-instance learning framework for performing the classification task using the proposed feature. Extensive experiments including comparison with a few state-of-art image classification approaches have been performed and the results suggest that the proposed approach is promising as it outperforms other methods by a large margin.",
author = "Ragav Venkatesan and Parag Chandakkar and Baoxin Li and Li, {Helen K.}",
year = "2012",
doi = "10.1109/EMBC.2012.6346216",
language = "English (US)",
isbn = "9781424441198",
pages = "1462--1465",
booktitle = "Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS",

}

TY - GEN

T1 - Classification of diabetic retinopathy images using multi-class multiple-instance learning based on color correlogram features

AU - Venkatesan, Ragav

AU - Chandakkar, Parag

AU - Li, Baoxin

AU - Li, Helen K.

PY - 2012

Y1 - 2012

N2 - All people with diabetes have the risk of developing diabetic retinopathy (DR), a vision-threatening complication. Early detection and timely treatment can reduce the occurrence of blindness due to DR. Computer-aided diagnosis has the potential benefit of improving the accuracy and speed in DR detection. This study is concerned with automatic classification of images with microaneurysm (MA) and neovascularization (NV), two important DR clinical findings. Together with normal images, this presents a 3-class classification problem. We propose a modified color auto-correlogram feature (AutoCC) with low dimensionality that is spectrally tuned towards DR images. Recognizing the fact that the images with or without MA or NV are generally different only in small, localized regions, we propose to employ a multi-class, multiple-instance learning framework for performing the classification task using the proposed feature. Extensive experiments including comparison with a few state-of-art image classification approaches have been performed and the results suggest that the proposed approach is promising as it outperforms other methods by a large margin.

AB - All people with diabetes have the risk of developing diabetic retinopathy (DR), a vision-threatening complication. Early detection and timely treatment can reduce the occurrence of blindness due to DR. Computer-aided diagnosis has the potential benefit of improving the accuracy and speed in DR detection. This study is concerned with automatic classification of images with microaneurysm (MA) and neovascularization (NV), two important DR clinical findings. Together with normal images, this presents a 3-class classification problem. We propose a modified color auto-correlogram feature (AutoCC) with low dimensionality that is spectrally tuned towards DR images. Recognizing the fact that the images with or without MA or NV are generally different only in small, localized regions, we propose to employ a multi-class, multiple-instance learning framework for performing the classification task using the proposed feature. Extensive experiments including comparison with a few state-of-art image classification approaches have been performed and the results suggest that the proposed approach is promising as it outperforms other methods by a large margin.

UR - http://www.scopus.com/inward/record.url?scp=84870795329&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84870795329&partnerID=8YFLogxK

U2 - 10.1109/EMBC.2012.6346216

DO - 10.1109/EMBC.2012.6346216

M3 - Conference contribution

C2 - 23366177

AN - SCOPUS:84870795329

SN - 9781424441198

SP - 1462

EP - 1465

BT - Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS

ER -