Deep Feature Transfer Learning in Combination with Traditional Features Predicts Survival Among Patients with Lung Adenocarcinoma

Authors

Rahul Paul, University of South Florida
Samuel H. Hawkings, University of South Florida
Matthew B. Schabath, H. Lee Moffitt Cancer Center
Robert J. Gilies, H. Lee Moffitt Cancer Center
Lawrence O. Hall, University of South FloridaFollow
Dmitry Goldgof, University of South FloridaFollow

Document Type

Article

Publication Date

12-2016

Keywords

pre-trained CNN, transfer learning, deep features, computed tomography, symmetric uncertainty, lung cancer, adenocarcinoma, deep neural network

Digital Object Identifier (DOI)

http://dx.doi.org/10.18383/j.tom.2016.00211

Abstract

Lung cancer is the most common cause of cancer-related deaths in the USA. It can be detected and diagnosed using computed tomography images. For an automated classifier, identifying predictive features from medical images is a key concern. Deep feature extraction using pretrained convolutional neural networks (CNNs) has recently been successfully applied in some image domains. Here, we applied a pretrained CNN to extract deep features from 40 computed tomography images, with contrast, of non-small cell adenocarcinoma lung cancer, and combined deep features with traditional image features and trained classifiers to predict short- and long-term survivors. We experimented with several pretrained CNNs and several feature selection strategies. The best previously reported accuracy when using traditional quantitative features was 77.5% (area under the curve [AUC], 0.712), which was achieved by a decision tree classifier. The best reported accuracy from transfer learning and deep features was 77.5% (AUC, 0.713) using a decision tree classifier. When extracted deep neural network features were combined with traditional quantitative features, we obtained an accuracy of 90% (AUC, 0.935) with the 5 best post-rectified linear unit features extracted from a vgg-f pretrained CNN and the 5 best traditional features. The best results were achieved with the symmetric uncertainty feature ranking algorithm followed by a random forests classifier.

Rights Information

This work is licensed under a Creative Commons Attribution 4.0 License.

Was this content written or created while at USF?

Yes

Citation / Publisher Attribution

TOMOGRAPHY, v. 2, no. 4, p. 388-395

Scholar Commons Citation

Paul, Rahul; Hawkings, Samuel H.; Schabath, Matthew B.; Gilies, Robert J.; Hall, Lawrence O.; and Goldgof, Dmitry, "Deep Feature Transfer Learning in Combination with Traditional Features Predicts Survival Among Patients with Lung Adenocarcinoma" (2016). Computer Science and Engineering Faculty Publications. 106.
https://digitalcommons.usf.edu/esb_facpub/106