An integrated texton and bag of words classifier for identifying anaplastic medulloblastomas

Joseph Galaro, Alexander R. Judkins, David Ellison, Jennifer Baccon, Anant Madabhushi

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

Abstract

In this paper we present a combined Bag of Words and texton based classifier for differentiating anaplastic and non-anaplastic medulloblastoma on digitized histopathology. The hypothesis behind this work is that histological image signatures may reflect different levels of aggressiveness of the disease and that texture based approaches can help discriminate between more aggressive and less aggressive phenotypes of medulloblastoma. The bag of words approach attempts to model the occurrence of differently expressed image features. In this work we choose to model the image features via textons which can quantitatively capture and model texture appearance in the images. The texton-based features, obtained via two methods, the Haar Wavelet responses and MR8 filter bank, provide spatial orientation and rotation invariant attributes. Applying these features to the bag of words framework yields textural representations that can be used in conjunction with a classifier (-nearest neighbor) or a content based image retrieval system. Over multiple runs of randomized cross validation, a -NN classifier in conjunction with Haar wavelets and the texton, bag of words approach yielded a mean classification accuracy of 80, an area under the precision recall curve of 87 and an area under the ROC curve of 83 in distinguishing between anaplastic and non-anaplastic medulloblastomas on a cohort of 36 patient studies.

Original languageEnglish (US)
Title of host publication33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011
Pages3443-3446
Number of pages4
DOIs
StatePublished - Dec 26 2011
Event33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011 - Boston, MA, United States
Duration: Aug 30 2011Sep 3 2011

Other

Other33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011
CountryUnited States
CityBoston, MA
Period8/30/119/3/11

Fingerprint

Medulloblastoma
Classifiers
Textures
Filter banks
Image retrieval
ROC Curve
Area Under Curve
Phenotype

All Science Journal Classification (ASJC) codes

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

Cite this

Galaro, J., Judkins, A. R., Ellison, D., Baccon, J., & Madabhushi, A. (2011). An integrated texton and bag of words classifier for identifying anaplastic medulloblastomas. In 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011 (pp. 3443-3446). [6090931] https://doi.org/10.1109/IEMBS.2011.6090931
Galaro, Joseph ; Judkins, Alexander R. ; Ellison, David ; Baccon, Jennifer ; Madabhushi, Anant. / An integrated texton and bag of words classifier for identifying anaplastic medulloblastomas. 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011. 2011. pp. 3443-3446
@inproceedings{5a40bfc2ec0849ebafd30469413d9798,
title = "An integrated texton and bag of words classifier for identifying anaplastic medulloblastomas",
abstract = "In this paper we present a combined Bag of Words and texton based classifier for differentiating anaplastic and non-anaplastic medulloblastoma on digitized histopathology. The hypothesis behind this work is that histological image signatures may reflect different levels of aggressiveness of the disease and that texture based approaches can help discriminate between more aggressive and less aggressive phenotypes of medulloblastoma. The bag of words approach attempts to model the occurrence of differently expressed image features. In this work we choose to model the image features via textons which can quantitatively capture and model texture appearance in the images. The texton-based features, obtained via two methods, the Haar Wavelet responses and MR8 filter bank, provide spatial orientation and rotation invariant attributes. Applying these features to the bag of words framework yields textural representations that can be used in conjunction with a classifier (-nearest neighbor) or a content based image retrieval system. Over multiple runs of randomized cross validation, a -NN classifier in conjunction with Haar wavelets and the texton, bag of words approach yielded a mean classification accuracy of 80, an area under the precision recall curve of 87 and an area under the ROC curve of 83 in distinguishing between anaplastic and non-anaplastic medulloblastomas on a cohort of 36 patient studies.",
author = "Joseph Galaro and Judkins, {Alexander R.} and David Ellison and Jennifer Baccon and Anant Madabhushi",
year = "2011",
month = "12",
day = "26",
doi = "10.1109/IEMBS.2011.6090931",
language = "English (US)",
isbn = "9781424441211",
pages = "3443--3446",
booktitle = "33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011",

}

Galaro, J, Judkins, AR, Ellison, D, Baccon, J & Madabhushi, A 2011, An integrated texton and bag of words classifier for identifying anaplastic medulloblastomas. in 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011., 6090931, pp. 3443-3446, 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011, Boston, MA, United States, 8/30/11. https://doi.org/10.1109/IEMBS.2011.6090931

An integrated texton and bag of words classifier for identifying anaplastic medulloblastomas. / Galaro, Joseph; Judkins, Alexander R.; Ellison, David; Baccon, Jennifer; Madabhushi, Anant.

33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011. 2011. p. 3443-3446 6090931.

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

TY - GEN

T1 - An integrated texton and bag of words classifier for identifying anaplastic medulloblastomas

AU - Galaro, Joseph

AU - Judkins, Alexander R.

AU - Ellison, David

AU - Baccon, Jennifer

AU - Madabhushi, Anant

PY - 2011/12/26

Y1 - 2011/12/26

N2 - In this paper we present a combined Bag of Words and texton based classifier for differentiating anaplastic and non-anaplastic medulloblastoma on digitized histopathology. The hypothesis behind this work is that histological image signatures may reflect different levels of aggressiveness of the disease and that texture based approaches can help discriminate between more aggressive and less aggressive phenotypes of medulloblastoma. The bag of words approach attempts to model the occurrence of differently expressed image features. In this work we choose to model the image features via textons which can quantitatively capture and model texture appearance in the images. The texton-based features, obtained via two methods, the Haar Wavelet responses and MR8 filter bank, provide spatial orientation and rotation invariant attributes. Applying these features to the bag of words framework yields textural representations that can be used in conjunction with a classifier (-nearest neighbor) or a content based image retrieval system. Over multiple runs of randomized cross validation, a -NN classifier in conjunction with Haar wavelets and the texton, bag of words approach yielded a mean classification accuracy of 80, an area under the precision recall curve of 87 and an area under the ROC curve of 83 in distinguishing between anaplastic and non-anaplastic medulloblastomas on a cohort of 36 patient studies.

AB - In this paper we present a combined Bag of Words and texton based classifier for differentiating anaplastic and non-anaplastic medulloblastoma on digitized histopathology. The hypothesis behind this work is that histological image signatures may reflect different levels of aggressiveness of the disease and that texture based approaches can help discriminate between more aggressive and less aggressive phenotypes of medulloblastoma. The bag of words approach attempts to model the occurrence of differently expressed image features. In this work we choose to model the image features via textons which can quantitatively capture and model texture appearance in the images. The texton-based features, obtained via two methods, the Haar Wavelet responses and MR8 filter bank, provide spatial orientation and rotation invariant attributes. Applying these features to the bag of words framework yields textural representations that can be used in conjunction with a classifier (-nearest neighbor) or a content based image retrieval system. Over multiple runs of randomized cross validation, a -NN classifier in conjunction with Haar wavelets and the texton, bag of words approach yielded a mean classification accuracy of 80, an area under the precision recall curve of 87 and an area under the ROC curve of 83 in distinguishing between anaplastic and non-anaplastic medulloblastomas on a cohort of 36 patient studies.

UR - http://www.scopus.com/inward/record.url?scp=84055176574&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84055176574&partnerID=8YFLogxK

U2 - 10.1109/IEMBS.2011.6090931

DO - 10.1109/IEMBS.2011.6090931

M3 - Conference contribution

SN - 9781424441211

SP - 3443

EP - 3446

BT - 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011

ER -

Galaro J, Judkins AR, Ellison D, Baccon J, Madabhushi A. An integrated texton and bag of words classifier for identifying anaplastic medulloblastomas. In 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS 2011. 2011. p. 3443-3446. 6090931 https://doi.org/10.1109/IEMBS.2011.6090931