Multi-level learning approach for prostate histopathology images classification

Shahnorbanun Sahran, Azizi Abdullah, Dheeb Albashish, Sawsan Jamil Abu-Taleb

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

1 Citation (Scopus)

Abstract

The classification of prostate histopathology images is concerned with the identification of multiple classes encompassing different grades of malignancy corresponding to the different observable textural patterns. To address the prostate multiclass classification problem, decomposition schemes are well-known techniques to solve the multiclass classification tasks. Among them two common binary approaches using one-versus-all (OVA) and one-versus-one (OVO) have gained a significant attention from the research community. However, in the case of OVO, the correlation between different classes is not considered as the multiclass problem is broken into multiple independent binary problems. On the other hand, OVA introduces an artificial class imbalance, which degrades the classification performance. In this paper, a new multiclass approach, named multi-level learning architecture (MLA), which handles the binary classification tasks in the multi-level strategy. It does so by taking the correlation between different classes and the domain knowledge into account. In addition, the proposed approach relies upon the 'divide and conquer' principle, and work by dividing each binary task into two separate tasks, strong and weak, based on the power of the samples in each binary task. In turn, this motivates the strong samples to produce the final prediction since they have more information about the considered task. Experiments on prostate histopathology images show that the MLA significantly outperforms existing OVO and OVA approaches when they applied on the ensemble framework. The results indicate the high effectiveness of the ensemble framework with MLA scheme in dealing with the prostate multiclass classification problem.

Original languageEnglish
Title of host publicationICIT 2017 - 8th International Conference on Information Technology, Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages947-955
Number of pages9
ISBN (Electronic)9781509063321
DOIs
Publication statusPublished - 20 Oct 2017
Event8th International Conference on Information Technology, ICIT 2017 - Amman, Jordan
Duration: 17 May 201718 May 2017

Other

Other8th International Conference on Information Technology, ICIT 2017
CountryJordan
CityAmman
Period17/5/1718/5/17

Fingerprint

Image classification
Prostate
Learning
Histopathology
Decomposition
Research
Neoplasms
Experiments

Keywords

  • ensemble classification
  • multiclasss classification
  • one-vs.-all
  • one-vs.-one
  • prostate grading

ASJC Scopus subject areas

  • Information Systems
  • Health Informatics
  • Information Systems and Management
  • Computer Networks and Communications
  • Computer Science Applications

Cite this

Sahran, S., Abdullah, A., Albashish, D., & Abu-Taleb, S. J. (2017). Multi-level learning approach for prostate histopathology images classification. In ICIT 2017 - 8th International Conference on Information Technology, Proceedings (pp. 947-955). [8079973] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICITECH.2017.8079973

Multi-level learning approach for prostate histopathology images classification. / Sahran, Shahnorbanun; Abdullah, Azizi; Albashish, Dheeb; Abu-Taleb, Sawsan Jamil.

ICIT 2017 - 8th International Conference on Information Technology, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 947-955 8079973.

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

Sahran, S, Abdullah, A, Albashish, D & Abu-Taleb, SJ 2017, Multi-level learning approach for prostate histopathology images classification. in ICIT 2017 - 8th International Conference on Information Technology, Proceedings., 8079973, Institute of Electrical and Electronics Engineers Inc., pp. 947-955, 8th International Conference on Information Technology, ICIT 2017, Amman, Jordan, 17/5/17. https://doi.org/10.1109/ICITECH.2017.8079973
Sahran S, Abdullah A, Albashish D, Abu-Taleb SJ. Multi-level learning approach for prostate histopathology images classification. In ICIT 2017 - 8th International Conference on Information Technology, Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. p. 947-955. 8079973 https://doi.org/10.1109/ICITECH.2017.8079973
Sahran, Shahnorbanun ; Abdullah, Azizi ; Albashish, Dheeb ; Abu-Taleb, Sawsan Jamil. / Multi-level learning approach for prostate histopathology images classification. ICIT 2017 - 8th International Conference on Information Technology, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 947-955
@inproceedings{d2c9aff0ccac4aebb796f1fe697febaa,
title = "Multi-level learning approach for prostate histopathology images classification",
abstract = "The classification of prostate histopathology images is concerned with the identification of multiple classes encompassing different grades of malignancy corresponding to the different observable textural patterns. To address the prostate multiclass classification problem, decomposition schemes are well-known techniques to solve the multiclass classification tasks. Among them two common binary approaches using one-versus-all (OVA) and one-versus-one (OVO) have gained a significant attention from the research community. However, in the case of OVO, the correlation between different classes is not considered as the multiclass problem is broken into multiple independent binary problems. On the other hand, OVA introduces an artificial class imbalance, which degrades the classification performance. In this paper, a new multiclass approach, named multi-level learning architecture (MLA), which handles the binary classification tasks in the multi-level strategy. It does so by taking the correlation between different classes and the domain knowledge into account. In addition, the proposed approach relies upon the 'divide and conquer' principle, and work by dividing each binary task into two separate tasks, strong and weak, based on the power of the samples in each binary task. In turn, this motivates the strong samples to produce the final prediction since they have more information about the considered task. Experiments on prostate histopathology images show that the MLA significantly outperforms existing OVO and OVA approaches when they applied on the ensemble framework. The results indicate the high effectiveness of the ensemble framework with MLA scheme in dealing with the prostate multiclass classification problem.",
keywords = "ensemble classification, multiclasss classification, one-vs.-all, one-vs.-one, prostate grading",
author = "Shahnorbanun Sahran and Azizi Abdullah and Dheeb Albashish and Abu-Taleb, {Sawsan Jamil}",
year = "2017",
month = "10",
day = "20",
doi = "10.1109/ICITECH.2017.8079973",
language = "English",
pages = "947--955",
booktitle = "ICIT 2017 - 8th International Conference on Information Technology, Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Multi-level learning approach for prostate histopathology images classification

AU - Sahran, Shahnorbanun

AU - Abdullah, Azizi

AU - Albashish, Dheeb

AU - Abu-Taleb, Sawsan Jamil

PY - 2017/10/20

Y1 - 2017/10/20

N2 - The classification of prostate histopathology images is concerned with the identification of multiple classes encompassing different grades of malignancy corresponding to the different observable textural patterns. To address the prostate multiclass classification problem, decomposition schemes are well-known techniques to solve the multiclass classification tasks. Among them two common binary approaches using one-versus-all (OVA) and one-versus-one (OVO) have gained a significant attention from the research community. However, in the case of OVO, the correlation between different classes is not considered as the multiclass problem is broken into multiple independent binary problems. On the other hand, OVA introduces an artificial class imbalance, which degrades the classification performance. In this paper, a new multiclass approach, named multi-level learning architecture (MLA), which handles the binary classification tasks in the multi-level strategy. It does so by taking the correlation between different classes and the domain knowledge into account. In addition, the proposed approach relies upon the 'divide and conquer' principle, and work by dividing each binary task into two separate tasks, strong and weak, based on the power of the samples in each binary task. In turn, this motivates the strong samples to produce the final prediction since they have more information about the considered task. Experiments on prostate histopathology images show that the MLA significantly outperforms existing OVO and OVA approaches when they applied on the ensemble framework. The results indicate the high effectiveness of the ensemble framework with MLA scheme in dealing with the prostate multiclass classification problem.

AB - The classification of prostate histopathology images is concerned with the identification of multiple classes encompassing different grades of malignancy corresponding to the different observable textural patterns. To address the prostate multiclass classification problem, decomposition schemes are well-known techniques to solve the multiclass classification tasks. Among them two common binary approaches using one-versus-all (OVA) and one-versus-one (OVO) have gained a significant attention from the research community. However, in the case of OVO, the correlation between different classes is not considered as the multiclass problem is broken into multiple independent binary problems. On the other hand, OVA introduces an artificial class imbalance, which degrades the classification performance. In this paper, a new multiclass approach, named multi-level learning architecture (MLA), which handles the binary classification tasks in the multi-level strategy. It does so by taking the correlation between different classes and the domain knowledge into account. In addition, the proposed approach relies upon the 'divide and conquer' principle, and work by dividing each binary task into two separate tasks, strong and weak, based on the power of the samples in each binary task. In turn, this motivates the strong samples to produce the final prediction since they have more information about the considered task. Experiments on prostate histopathology images show that the MLA significantly outperforms existing OVO and OVA approaches when they applied on the ensemble framework. The results indicate the high effectiveness of the ensemble framework with MLA scheme in dealing with the prostate multiclass classification problem.

KW - ensemble classification

KW - multiclasss classification

KW - one-vs.-all

KW - one-vs.-one

KW - prostate grading

UR - http://www.scopus.com/inward/record.url?scp=85039984465&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85039984465&partnerID=8YFLogxK

U2 - 10.1109/ICITECH.2017.8079973

DO - 10.1109/ICITECH.2017.8079973

M3 - Conference contribution

SP - 947

EP - 955

BT - ICIT 2017 - 8th International Conference on Information Technology, Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -