Pulmonary nodule diagnosis using dual-modal supervised autoencoder based on extreme learning machine

Yan Qiang, Lei Ge, Xin Zhao, Xiaolong Zhang, Xiaoxian Tang

Research output: Contribution to journalReview article

5 Citations (Scopus)

Abstract

In recent years, deep learning techniques have been applied to the diagnosis of pulmonary nodules. In order to improve the pulmonary nodule diagnostic performance effectively, we propose a novel pulmonary nodule diagnosis method using dual-modal deep supervised autoencoder based on extreme learning machine for which discriminative features are automatically learnt from the input data. The network is fed with nodule images in pairs obtained from computed tomography and positron emission tomography respectively. For each pair image, the high-level discriminative features of nodules in computed tomography and positron emission tomography are extracted from stacked supervised autoencoder layers. The outputs of the proposed architecture are combined using an ideal fusion method to get the final classification. In the experiments, 5-fold cross-validation method is used to validate the proposed method on 1,600 pulmonary nodule images and our method reaches high-classification sensitivities of 91.75% at 1.58 false positives per scan. Meanwhile, compared with other deep learning diagnosis methods, our method achieves better discriminative results and is highly suited to be used for pulmonary nodule diagnosis.

Original languageEnglish (US)
Article numbere12224
JournalExpert Systems
Volume34
Issue number6
DOIs
StatePublished - Dec 1 2017

Fingerprint

Extreme Learning Machine
Nodule
Learning systems
Positron emission tomography
Tomography
Positron Emission Tomography
Computed Tomography
Fusion reactions
Dual Method
False Positive
Cross-validation
Diagnostics
Fusion
Fold
Experiments
Output
Deep learning

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Theoretical Computer Science
  • Computational Theory and Mathematics
  • Artificial Intelligence

Cite this

@article{219c8dc7bf434f208362438bfe676022,
title = "Pulmonary nodule diagnosis using dual-modal supervised autoencoder based on extreme learning machine",
abstract = "In recent years, deep learning techniques have been applied to the diagnosis of pulmonary nodules. In order to improve the pulmonary nodule diagnostic performance effectively, we propose a novel pulmonary nodule diagnosis method using dual-modal deep supervised autoencoder based on extreme learning machine for which discriminative features are automatically learnt from the input data. The network is fed with nodule images in pairs obtained from computed tomography and positron emission tomography respectively. For each pair image, the high-level discriminative features of nodules in computed tomography and positron emission tomography are extracted from stacked supervised autoencoder layers. The outputs of the proposed architecture are combined using an ideal fusion method to get the final classification. In the experiments, 5-fold cross-validation method is used to validate the proposed method on 1,600 pulmonary nodule images and our method reaches high-classification sensitivities of 91.75{\%} at 1.58 false positives per scan. Meanwhile, compared with other deep learning diagnosis methods, our method achieves better discriminative results and is highly suited to be used for pulmonary nodule diagnosis.",
author = "Yan Qiang and Lei Ge and Xin Zhao and Xiaolong Zhang and Xiaoxian Tang",
year = "2017",
month = "12",
day = "1",
doi = "10.1111/exsy.12224",
language = "English (US)",
volume = "34",
journal = "Expert Systems",
issn = "0266-4720",
publisher = "Wiley-Blackwell",
number = "6",

}

Pulmonary nodule diagnosis using dual-modal supervised autoencoder based on extreme learning machine. / Qiang, Yan; Ge, Lei; Zhao, Xin; Zhang, Xiaolong; Tang, Xiaoxian.

In: Expert Systems, Vol. 34, No. 6, e12224, 01.12.2017.

Research output: Contribution to journalReview article

TY - JOUR

T1 - Pulmonary nodule diagnosis using dual-modal supervised autoencoder based on extreme learning machine

AU - Qiang, Yan

AU - Ge, Lei

AU - Zhao, Xin

AU - Zhang, Xiaolong

AU - Tang, Xiaoxian

PY - 2017/12/1

Y1 - 2017/12/1

N2 - In recent years, deep learning techniques have been applied to the diagnosis of pulmonary nodules. In order to improve the pulmonary nodule diagnostic performance effectively, we propose a novel pulmonary nodule diagnosis method using dual-modal deep supervised autoencoder based on extreme learning machine for which discriminative features are automatically learnt from the input data. The network is fed with nodule images in pairs obtained from computed tomography and positron emission tomography respectively. For each pair image, the high-level discriminative features of nodules in computed tomography and positron emission tomography are extracted from stacked supervised autoencoder layers. The outputs of the proposed architecture are combined using an ideal fusion method to get the final classification. In the experiments, 5-fold cross-validation method is used to validate the proposed method on 1,600 pulmonary nodule images and our method reaches high-classification sensitivities of 91.75% at 1.58 false positives per scan. Meanwhile, compared with other deep learning diagnosis methods, our method achieves better discriminative results and is highly suited to be used for pulmonary nodule diagnosis.

AB - In recent years, deep learning techniques have been applied to the diagnosis of pulmonary nodules. In order to improve the pulmonary nodule diagnostic performance effectively, we propose a novel pulmonary nodule diagnosis method using dual-modal deep supervised autoencoder based on extreme learning machine for which discriminative features are automatically learnt from the input data. The network is fed with nodule images in pairs obtained from computed tomography and positron emission tomography respectively. For each pair image, the high-level discriminative features of nodules in computed tomography and positron emission tomography are extracted from stacked supervised autoencoder layers. The outputs of the proposed architecture are combined using an ideal fusion method to get the final classification. In the experiments, 5-fold cross-validation method is used to validate the proposed method on 1,600 pulmonary nodule images and our method reaches high-classification sensitivities of 91.75% at 1.58 false positives per scan. Meanwhile, compared with other deep learning diagnosis methods, our method achieves better discriminative results and is highly suited to be used for pulmonary nodule diagnosis.

UR - http://www.scopus.com/inward/record.url?scp=85024867178&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85024867178&partnerID=8YFLogxK

U2 - 10.1111/exsy.12224

DO - 10.1111/exsy.12224

M3 - Review article

AN - SCOPUS:85024867178

VL - 34

JO - Expert Systems

JF - Expert Systems

SN - 0266-4720

IS - 6

M1 - e12224

ER -