Predicting malignant nodules by fusing deep features with classical radiomics features

Rahul Paul; Samuel Hawkins; Matthew B. Schabath; Robert J. Gillies; Lawrence O. Hall; Dmitry B. Goldgof

doi:10.1117/1.JMI.5.1.011021

21 March 2018 Predicting malignant nodules by fusing deep features with classical radiomics features

Rahul Paul, Samuel Hawkins, Matthew B. Schabath, Robert J. Gillies, Lawrence O. Hall, Dmitry B. Goldgof

Author Affiliations +

Journal of Medical Imaging, Vol. 5, Issue 1, 011021 (March 2018). https://doi.org/10.1117/1.JMI.5.1.011021

Abstract

Lung cancer has a high incidence and mortality rate. Early detection and diagnosis of lung cancers is best achieved with low-dose computed tomography (CT). Classical radiomics features extracted from lung CT images have been shown as able to predict cancer incidence and prognosis. With the advancement of deep learning and convolutional neural networks (CNNs), deep features can be identified to analyze lung CTs for prognosis prediction and diagnosis. Due to a limited number of available images in the medical field, the transfer learning concept can be helpful. Using subsets of participants from the National Lung Screening Trial (NLST), we utilized a transfer learning approach to differentiate lung cancer nodules versus positive controls. We experimented with three different pretrained CNNs for extracting deep features and used five different classifiers. Experiments were also conducted with deep features from different color channels of a pretrained CNN. Selected deep features were combined with radiomics features. A CNN was designed and trained. Combinations of features from pretrained, CNNs trained on NLST data, and classical radiomics were used to build classifiers. The best accuracy (76.79%) was obtained using feature combinations. An area under the receiver operating characteristic curve of 0.87 was obtained using a CNN trained on an augmented NLST data cohort.

Citation Download Citation

Rahul Paul, Samuel Hawkins, Matthew B. Schabath, Robert J. Gillies, Lawrence O. Hall, and Dmitry B. Goldgof "Predicting malignant nodules by fusing deep features with classical radiomics features," Journal of Medical Imaging 5(1), 011021 (21 March 2018). https://doi.org/10.1117/1.JMI.5.1.011021

Received: 6 July 2017; Accepted: 19 February 2018; Published: 21 March 2018

ACCESS THE FULL ARTICLE

JOURNAL ARTICLE
11 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

CITATIONS

Cited by 75 scholarly publications and 4 patents.

Explore citations on Lens.org

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Lung cancer

Lung

Feature extraction

Cancer

RGB color model

Control systems

Computed tomography

Show All Keywords

RELATED CONTENT

A neural network approach to lung nodule segmentation
Proceedings of SPIE (March 21 2016)

Towards deep radiomics nodule malignancy prediction using CNNs on...
Proceedings of SPIE (March 13 2019)

Contextual convolutional neural networks for lung nodule classification using Gaussian...
Proceedings of SPIE (March 03 2017)

An automatic end to end pipeline for CT image based...
Proceedings of SPIE (March 15 2019)

Computer aided classification of pulmonary nodules in surrounding and internal...
Proceedings of SPIE (May 09 2002)

Characterization of pulmonary nodules on computer tomography (CT) scans ...
Proceedings of SPIE (March 05 2007)

Algorithm for lung cancer detection based on PET/CT images
Proceedings of SPIE (February 27 2009)

Subscribe to Digital Library

Receive Erratum Email Alert

Show All Keywords

Keywords/Phrases

Search In:

Publication Years