A novel fused convolutional neural network for biomedical image classification

Shuchao Pang, Anan Du, Mehmet Orgun, Zhezhou Yu

Research output: Contribution to journalArticleResearchpeer-review

Abstract

With the advent of biomedical imaging technology, the number of captured and stored biomedical images is rapidly increasing day by day in hospitals, imaging laboratories and biomedical institutions. Therefore, more robust biomedical image analysis technology is needed to meet the requirement of the diagnosis and classification of various kinds of diseases using biomedical images. However, the current biomedical image classification methods and general non-biomedical image classifiers cannot extract more compact biomedical image features or capture the tiny differences between similar images with different types of diseases from the same category. In this paper, we propose a novel fused convolutional neural network to develop a more accurate and highly efficient classifier for biomedical images, which combines shallow layer features and deep layer features from the proposed deep neural network architecture. In the analysis, it was observed that the shallow layers provided more detailed local features, which could distinguish different diseases in the same category, while the deep layers could convey more high-level semantic information used to classify the diseases among the various categories. A detailed comparison of our approach with traditional classification algorithms and popular deep classifiers across several public biomedical image datasets showed the superior performance of our proposed method for biomedical image classification. In addition, we also evaluated the performance of our method in modality classification of medical images using the ImageCLEFmed dataset. [Figure not available: see fulltext.].

LanguageEnglish
Pages107-121
Number of pages15
JournalMedical & Biological Engineering & Computing
Volume57
Issue number1
DOIs
Publication statusPublished - Jan 2019

Fingerprint

Image classification
Neural networks
Classifiers
Imaging techniques
Network architecture
Image analysis
Semantics

Keywords

  • biomedical image classification
  • convolutional neural networks
  • deep learning
  • deep feature
  • shallow feature

Cite this

@article{5a25dd22a3c44b2aba97ec0fa6cdaca5,
title = "A novel fused convolutional neural network for biomedical image classification",
abstract = "With the advent of biomedical imaging technology, the number of captured and stored biomedical images is rapidly increasing day by day in hospitals, imaging laboratories and biomedical institutions. Therefore, more robust biomedical image analysis technology is needed to meet the requirement of the diagnosis and classification of various kinds of diseases using biomedical images. However, the current biomedical image classification methods and general non-biomedical image classifiers cannot extract more compact biomedical image features or capture the tiny differences between similar images with different types of diseases from the same category. In this paper, we propose a novel fused convolutional neural network to develop a more accurate and highly efficient classifier for biomedical images, which combines shallow layer features and deep layer features from the proposed deep neural network architecture. In the analysis, it was observed that the shallow layers provided more detailed local features, which could distinguish different diseases in the same category, while the deep layers could convey more high-level semantic information used to classify the diseases among the various categories. A detailed comparison of our approach with traditional classification algorithms and popular deep classifiers across several public biomedical image datasets showed the superior performance of our proposed method for biomedical image classification. In addition, we also evaluated the performance of our method in modality classification of medical images using the ImageCLEFmed dataset. [Figure not available: see fulltext.].",
keywords = "biomedical image classification, convolutional neural networks, deep learning, deep feature, shallow feature",
author = "Shuchao Pang and Anan Du and Mehmet Orgun and Zhezhou Yu",
year = "2019",
month = "1",
doi = "10.1007/s11517-018-1819-y",
language = "English",
volume = "57",
pages = "107--121",
journal = "Medical & Biological Engineering & Computing",
issn = "0140-0118",
publisher = "Springer, Springer Nature",
number = "1",

}

A novel fused convolutional neural network for biomedical image classification. / Pang, Shuchao; Du, Anan; Orgun, Mehmet; Yu, Zhezhou.

In: Medical & Biological Engineering & Computing, Vol. 57, No. 1, 01.2019, p. 107-121.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - A novel fused convolutional neural network for biomedical image classification

AU - Pang, Shuchao

AU - Du, Anan

AU - Orgun, Mehmet

AU - Yu, Zhezhou

PY - 2019/1

Y1 - 2019/1

N2 - With the advent of biomedical imaging technology, the number of captured and stored biomedical images is rapidly increasing day by day in hospitals, imaging laboratories and biomedical institutions. Therefore, more robust biomedical image analysis technology is needed to meet the requirement of the diagnosis and classification of various kinds of diseases using biomedical images. However, the current biomedical image classification methods and general non-biomedical image classifiers cannot extract more compact biomedical image features or capture the tiny differences between similar images with different types of diseases from the same category. In this paper, we propose a novel fused convolutional neural network to develop a more accurate and highly efficient classifier for biomedical images, which combines shallow layer features and deep layer features from the proposed deep neural network architecture. In the analysis, it was observed that the shallow layers provided more detailed local features, which could distinguish different diseases in the same category, while the deep layers could convey more high-level semantic information used to classify the diseases among the various categories. A detailed comparison of our approach with traditional classification algorithms and popular deep classifiers across several public biomedical image datasets showed the superior performance of our proposed method for biomedical image classification. In addition, we also evaluated the performance of our method in modality classification of medical images using the ImageCLEFmed dataset. [Figure not available: see fulltext.].

AB - With the advent of biomedical imaging technology, the number of captured and stored biomedical images is rapidly increasing day by day in hospitals, imaging laboratories and biomedical institutions. Therefore, more robust biomedical image analysis technology is needed to meet the requirement of the diagnosis and classification of various kinds of diseases using biomedical images. However, the current biomedical image classification methods and general non-biomedical image classifiers cannot extract more compact biomedical image features or capture the tiny differences between similar images with different types of diseases from the same category. In this paper, we propose a novel fused convolutional neural network to develop a more accurate and highly efficient classifier for biomedical images, which combines shallow layer features and deep layer features from the proposed deep neural network architecture. In the analysis, it was observed that the shallow layers provided more detailed local features, which could distinguish different diseases in the same category, while the deep layers could convey more high-level semantic information used to classify the diseases among the various categories. A detailed comparison of our approach with traditional classification algorithms and popular deep classifiers across several public biomedical image datasets showed the superior performance of our proposed method for biomedical image classification. In addition, we also evaluated the performance of our method in modality classification of medical images using the ImageCLEFmed dataset. [Figure not available: see fulltext.].

KW - biomedical image classification

KW - convolutional neural networks

KW - deep learning

KW - deep feature

KW - shallow feature

UR - http://www.scopus.com/inward/record.url?scp=85049792899&partnerID=8YFLogxK

U2 - 10.1007/s11517-018-1819-y

DO - 10.1007/s11517-018-1819-y

M3 - Article

VL - 57

SP - 107

EP - 121

JO - Medical & Biological Engineering & Computing

T2 - Medical & Biological Engineering & Computing

JF - Medical & Biological Engineering & Computing

SN - 0140-0118

IS - 1

ER -