Breast cancer diagnosis in DCE-MRI using mixture ensemble of convolutional neural networks

(2017) Breast cancer diagnosis in DCE-MRI using mixture ensemble of convolutional neural networks. Pattern Recognition. pp. 381-390. ISSN 0031-3203

Full text not available from this repository.

Abstract

This work addresses a novel computer-aided diagnosis (CAD) system in breast dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). The CAD system is designed based on a mixture ensemble of convolutional neural networks (ME-CNN) to discriminate between benign and malignant breast tumors. The ME-CNN is a modular and image-based ensemble, which can stochastically partition the high dimensional image space through simultaneous and competitive learning of its modules. The proposed system was assessed on our database of 112 DCE-MRI studies including solid breast masses, using a wide range of classification measures. The ME-CNN model composed of three CNN experts and one convolutional gating network achieves an accuracy of 96.39, a sensitivity of 97.73 and a specificity of 94.87. The experimental results also show that it has competitive classification performances compared to three existing single-classifier methods and two convolutional ensemble methods. The proposed ME CNN model could provide an effective tool for radiologists to analyse breast DCE-MRI images. (C) 2017 Elsevier Ltd. All rights reserved.

Item Type: Article
Keywords: breast cancer dce-mri convolutional neural networks mixture ensemble of experts cad systems computer-aided-diagnosis images classification segmentation recognition features lesion prediction morphology nuclei
Divisions: Other
School of Advanced Technologies in Medicine > Department of Bioelectrics and Biomedical Engineering
Page Range: pp. 381-390
Journal or Publication Title: Pattern Recognition
Journal Index: ISI
Volume: 72
Identification Number: https://doi.org/10.1016/j.patcog.2017.08.004
ISSN: 0031-3203
Depositing User: مهندس مهدی شریفی
URI: http://eprints.mui.ac.ir/id/eprint/73

Actions (login required)

View Item View Item