Development and evaluation of an automated histology classification system for veterinary pathology

Arthur Hattel, Vishal Monga, Umamahesh Srinivas, Jim Gillespie, Jason Brooks, Jenny Fisher, Bhushan Jayarao

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

A 2-stage algorithmic framework was developed to automatically classify digitized photomicrographs of tissues obtained from bovine liver, lung, spleen, and kidney into different histologic categories. The categories included normal tissue, acute necrosis, and inflammation (acute suppurative; chronic). In the current study, a total of 60 images per category (normal; acute necrosis; acute suppurative inflammation) were obtained from liver samples, 60 images per category (normal; acute suppurative inflammation) were obtained from spleen and lung samples, and 60 images per category (normal; chronic inflammation) were obtained from kidney samples. An automated support vector machine (SVM) classifier was trained to assign each test image to a specific category. Using 10 training images/category/organ, 40 test images/category/organ were examined. Employing confusion matrices to represent category-specific classification accuracy, the classifier-attained accuracies were found to be in the 74-90% range. The same set of test images was evaluated using a SVM classifier trained on 20 images/category/organ. The average classification accuracies were noted to be in the 84-95% range. The accuracy in correctly identifying normal tissue and specific tissue lesions was markedly improved by a small increase in the number of training images. The preliminary results from the study indicate the importance and potential use of automated image classification systems in the histologic identification of normal tissues and specific tissue lesions.

Original languageEnglish (US)
Pages (from-to)765-769
Number of pages5
JournalJournal of Veterinary Diagnostic Investigation
Volume25
Issue number6
DOIs
StatePublished - Nov 1 2013

Fingerprint

Veterinary Pathology
histology
Histology
inflammation
Inflammation
lesions (animal)
necrosis
spleen
Necrosis
Spleen
lungs
kidneys
Kidney
Lung
liver
Liver
testing
sampling
tissues
cattle

All Science Journal Classification (ASJC) codes

  • veterinary(all)

Cite this

@article{bf7cff532ff842998788afe4f2627f4c,
title = "Development and evaluation of an automated histology classification system for veterinary pathology",
abstract = "A 2-stage algorithmic framework was developed to automatically classify digitized photomicrographs of tissues obtained from bovine liver, lung, spleen, and kidney into different histologic categories. The categories included normal tissue, acute necrosis, and inflammation (acute suppurative; chronic). In the current study, a total of 60 images per category (normal; acute necrosis; acute suppurative inflammation) were obtained from liver samples, 60 images per category (normal; acute suppurative inflammation) were obtained from spleen and lung samples, and 60 images per category (normal; chronic inflammation) were obtained from kidney samples. An automated support vector machine (SVM) classifier was trained to assign each test image to a specific category. Using 10 training images/category/organ, 40 test images/category/organ were examined. Employing confusion matrices to represent category-specific classification accuracy, the classifier-attained accuracies were found to be in the 74-90{\%} range. The same set of test images was evaluated using a SVM classifier trained on 20 images/category/organ. The average classification accuracies were noted to be in the 84-95{\%} range. The accuracy in correctly identifying normal tissue and specific tissue lesions was markedly improved by a small increase in the number of training images. The preliminary results from the study indicate the importance and potential use of automated image classification systems in the histologic identification of normal tissues and specific tissue lesions.",
author = "Arthur Hattel and Vishal Monga and Umamahesh Srinivas and Jim Gillespie and Jason Brooks and Jenny Fisher and Bhushan Jayarao",
year = "2013",
month = "11",
day = "1",
doi = "10.1177/1040638713506901",
language = "English (US)",
volume = "25",
pages = "765--769",
journal = "Journal of Veterinary Diagnostic Investigation",
issn = "1040-6387",
publisher = "American Association of Veterinary Laboratory Diagnosticians",
number = "6",

}

Development and evaluation of an automated histology classification system for veterinary pathology. / Hattel, Arthur; Monga, Vishal; Srinivas, Umamahesh; Gillespie, Jim; Brooks, Jason; Fisher, Jenny; Jayarao, Bhushan.

In: Journal of Veterinary Diagnostic Investigation, Vol. 25, No. 6, 01.11.2013, p. 765-769.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Development and evaluation of an automated histology classification system for veterinary pathology

AU - Hattel, Arthur

AU - Monga, Vishal

AU - Srinivas, Umamahesh

AU - Gillespie, Jim

AU - Brooks, Jason

AU - Fisher, Jenny

AU - Jayarao, Bhushan

PY - 2013/11/1

Y1 - 2013/11/1

N2 - A 2-stage algorithmic framework was developed to automatically classify digitized photomicrographs of tissues obtained from bovine liver, lung, spleen, and kidney into different histologic categories. The categories included normal tissue, acute necrosis, and inflammation (acute suppurative; chronic). In the current study, a total of 60 images per category (normal; acute necrosis; acute suppurative inflammation) were obtained from liver samples, 60 images per category (normal; acute suppurative inflammation) were obtained from spleen and lung samples, and 60 images per category (normal; chronic inflammation) were obtained from kidney samples. An automated support vector machine (SVM) classifier was trained to assign each test image to a specific category. Using 10 training images/category/organ, 40 test images/category/organ were examined. Employing confusion matrices to represent category-specific classification accuracy, the classifier-attained accuracies were found to be in the 74-90% range. The same set of test images was evaluated using a SVM classifier trained on 20 images/category/organ. The average classification accuracies were noted to be in the 84-95% range. The accuracy in correctly identifying normal tissue and specific tissue lesions was markedly improved by a small increase in the number of training images. The preliminary results from the study indicate the importance and potential use of automated image classification systems in the histologic identification of normal tissues and specific tissue lesions.

AB - A 2-stage algorithmic framework was developed to automatically classify digitized photomicrographs of tissues obtained from bovine liver, lung, spleen, and kidney into different histologic categories. The categories included normal tissue, acute necrosis, and inflammation (acute suppurative; chronic). In the current study, a total of 60 images per category (normal; acute necrosis; acute suppurative inflammation) were obtained from liver samples, 60 images per category (normal; acute suppurative inflammation) were obtained from spleen and lung samples, and 60 images per category (normal; chronic inflammation) were obtained from kidney samples. An automated support vector machine (SVM) classifier was trained to assign each test image to a specific category. Using 10 training images/category/organ, 40 test images/category/organ were examined. Employing confusion matrices to represent category-specific classification accuracy, the classifier-attained accuracies were found to be in the 74-90% range. The same set of test images was evaluated using a SVM classifier trained on 20 images/category/organ. The average classification accuracies were noted to be in the 84-95% range. The accuracy in correctly identifying normal tissue and specific tissue lesions was markedly improved by a small increase in the number of training images. The preliminary results from the study indicate the importance and potential use of automated image classification systems in the histologic identification of normal tissues and specific tissue lesions.

UR - http://www.scopus.com/inward/record.url?scp=84887326417&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84887326417&partnerID=8YFLogxK

U2 - 10.1177/1040638713506901

DO - 10.1177/1040638713506901

M3 - Article

C2 - 24153030

AN - SCOPUS:84887326417

VL - 25

SP - 765

EP - 769

JO - Journal of Veterinary Diagnostic Investigation

JF - Journal of Veterinary Diagnostic Investigation

SN - 1040-6387

IS - 6

ER -