Histopathological diagnosis for viable and non-viable tumor prediction for osteosarcoma using convolutional neural network

Rashika Mishra, Ovidiu Daescu, Patrick J Leavey, Dinesh Rakheja, Anita L Sengupta

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

8 Citations (Scopus)

Abstract

Pathologists often deal with high complexity and sometimes disagreement over Osteosarcoma tumor classification due to cellular heterogeneity in the dataset. Segmentation and classification of histology tissue in H&E stained tumor image datasets is challenging due to intra-class variations and inter-class similarity, crowded context, and noisy data. In recent years, deep learning approaches have led to encouraging results in breast cancer and prostate cancer analysis. In this paper, we propose a Convolutional neural network (CNN) as a tool to improve efficiency and accuracy of Osteosarcoma tumor classification into tumor classes (viable tumor, necrosis) vs non-tumor. The proposed CNN architecture contains five learned layers: three convolutional layers interspersed with max pooling layers for feature extraction and two fully-connected layers with data augmentation strategies to boost performance. We conclude that the use of neural network can assure high accuracy and efficiency in Osteosarcoma classification.

Original languageEnglish (US)
Title of host publicationBioinformatics Research and Applications - 13th International Symposium, ISBRA 2017, Proceedings
PublisherSpringer Verlag
Pages12-23
Number of pages12
Volume10330 LNBI
ISBN (Print)9783319595740
DOIs
StatePublished - 2017
Event13th International Symposium on Bioinformatics Research and Applications, ISBRA 2017 - Honolulu, United States
Duration: May 29 2017Jun 2 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10330 LNBI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other13th International Symposium on Bioinformatics Research and Applications, ISBRA 2017
CountryUnited States
CityHonolulu
Period5/29/176/2/17

Fingerprint

Tumors
Tumor
Neural Networks
Neural networks
Prediction
Data Augmentation
Histology
Necrosis
Prostate Cancer
Pooling
Noisy Data
Network Architecture
Network architecture
Breast Cancer
Feature Extraction
High Efficiency
Feature extraction
High Accuracy
Segmentation
Tissue

Keywords

  • Convolutional neural network
  • Histology image analysis
  • Osteosarcoma

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Mishra, R., Daescu, O., Leavey, P. J., Rakheja, D., & Sengupta, A. L. (2017). Histopathological diagnosis for viable and non-viable tumor prediction for osteosarcoma using convolutional neural network. In Bioinformatics Research and Applications - 13th International Symposium, ISBRA 2017, Proceedings (Vol. 10330 LNBI, pp. 12-23). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10330 LNBI). Springer Verlag. https://doi.org/10.1007/978-3-319-59575-7_2

Histopathological diagnosis for viable and non-viable tumor prediction for osteosarcoma using convolutional neural network. / Mishra, Rashika; Daescu, Ovidiu; Leavey, Patrick J; Rakheja, Dinesh; Sengupta, Anita L.

Bioinformatics Research and Applications - 13th International Symposium, ISBRA 2017, Proceedings. Vol. 10330 LNBI Springer Verlag, 2017. p. 12-23 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10330 LNBI).

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

Mishra, R, Daescu, O, Leavey, PJ, Rakheja, D & Sengupta, AL 2017, Histopathological diagnosis for viable and non-viable tumor prediction for osteosarcoma using convolutional neural network. in Bioinformatics Research and Applications - 13th International Symposium, ISBRA 2017, Proceedings. vol. 10330 LNBI, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10330 LNBI, Springer Verlag, pp. 12-23, 13th International Symposium on Bioinformatics Research and Applications, ISBRA 2017, Honolulu, United States, 5/29/17. https://doi.org/10.1007/978-3-319-59575-7_2
Mishra R, Daescu O, Leavey PJ, Rakheja D, Sengupta AL. Histopathological diagnosis for viable and non-viable tumor prediction for osteosarcoma using convolutional neural network. In Bioinformatics Research and Applications - 13th International Symposium, ISBRA 2017, Proceedings. Vol. 10330 LNBI. Springer Verlag. 2017. p. 12-23. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-59575-7_2
Mishra, Rashika ; Daescu, Ovidiu ; Leavey, Patrick J ; Rakheja, Dinesh ; Sengupta, Anita L. / Histopathological diagnosis for viable and non-viable tumor prediction for osteosarcoma using convolutional neural network. Bioinformatics Research and Applications - 13th International Symposium, ISBRA 2017, Proceedings. Vol. 10330 LNBI Springer Verlag, 2017. pp. 12-23 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{ce61874341a3435792c02f265d32b93d,
title = "Histopathological diagnosis for viable and non-viable tumor prediction for osteosarcoma using convolutional neural network",
abstract = "Pathologists often deal with high complexity and sometimes disagreement over Osteosarcoma tumor classification due to cellular heterogeneity in the dataset. Segmentation and classification of histology tissue in H&E stained tumor image datasets is challenging due to intra-class variations and inter-class similarity, crowded context, and noisy data. In recent years, deep learning approaches have led to encouraging results in breast cancer and prostate cancer analysis. In this paper, we propose a Convolutional neural network (CNN) as a tool to improve efficiency and accuracy of Osteosarcoma tumor classification into tumor classes (viable tumor, necrosis) vs non-tumor. The proposed CNN architecture contains five learned layers: three convolutional layers interspersed with max pooling layers for feature extraction and two fully-connected layers with data augmentation strategies to boost performance. We conclude that the use of neural network can assure high accuracy and efficiency in Osteosarcoma classification.",
keywords = "Convolutional neural network, Histology image analysis, Osteosarcoma",
author = "Rashika Mishra and Ovidiu Daescu and Leavey, {Patrick J} and Dinesh Rakheja and Sengupta, {Anita L}",
year = "2017",
doi = "10.1007/978-3-319-59575-7_2",
language = "English (US)",
isbn = "9783319595740",
volume = "10330 LNBI",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "12--23",
booktitle = "Bioinformatics Research and Applications - 13th International Symposium, ISBRA 2017, Proceedings",

}

TY - GEN

T1 - Histopathological diagnosis for viable and non-viable tumor prediction for osteosarcoma using convolutional neural network

AU - Mishra, Rashika

AU - Daescu, Ovidiu

AU - Leavey, Patrick J

AU - Rakheja, Dinesh

AU - Sengupta, Anita L

PY - 2017

Y1 - 2017

N2 - Pathologists often deal with high complexity and sometimes disagreement over Osteosarcoma tumor classification due to cellular heterogeneity in the dataset. Segmentation and classification of histology tissue in H&E stained tumor image datasets is challenging due to intra-class variations and inter-class similarity, crowded context, and noisy data. In recent years, deep learning approaches have led to encouraging results in breast cancer and prostate cancer analysis. In this paper, we propose a Convolutional neural network (CNN) as a tool to improve efficiency and accuracy of Osteosarcoma tumor classification into tumor classes (viable tumor, necrosis) vs non-tumor. The proposed CNN architecture contains five learned layers: three convolutional layers interspersed with max pooling layers for feature extraction and two fully-connected layers with data augmentation strategies to boost performance. We conclude that the use of neural network can assure high accuracy and efficiency in Osteosarcoma classification.

AB - Pathologists often deal with high complexity and sometimes disagreement over Osteosarcoma tumor classification due to cellular heterogeneity in the dataset. Segmentation and classification of histology tissue in H&E stained tumor image datasets is challenging due to intra-class variations and inter-class similarity, crowded context, and noisy data. In recent years, deep learning approaches have led to encouraging results in breast cancer and prostate cancer analysis. In this paper, we propose a Convolutional neural network (CNN) as a tool to improve efficiency and accuracy of Osteosarcoma tumor classification into tumor classes (viable tumor, necrosis) vs non-tumor. The proposed CNN architecture contains five learned layers: three convolutional layers interspersed with max pooling layers for feature extraction and two fully-connected layers with data augmentation strategies to boost performance. We conclude that the use of neural network can assure high accuracy and efficiency in Osteosarcoma classification.

KW - Convolutional neural network

KW - Histology image analysis

KW - Osteosarcoma

UR - http://www.scopus.com/inward/record.url?scp=85020721422&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85020721422&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-59575-7_2

DO - 10.1007/978-3-319-59575-7_2

M3 - Conference contribution

AN - SCOPUS:85020721422

SN - 9783319595740

VL - 10330 LNBI

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 12

EP - 23

BT - Bioinformatics Research and Applications - 13th International Symposium, ISBRA 2017, Proceedings

PB - Springer Verlag

ER -