Predicting malignant nodules by fusing deep features with classical radiomics features
Autor: | Samuel H. Hawkins, Rahul Paul, Lawrence O. Hall, Dmitry B. Goldgof, Robert J. Gillies, Matthew B. Schabath |
---|---|
Rok vydání: | 2017 |
Předmět: |
Artificial neural network
Contextual image classification Receiver operating characteristic business.industry Deep learning Feature extraction Pattern recognition Convolutional neural network 030218 nuclear medicine & medical imaging 03 medical and health sciences 0302 clinical medicine Feature (computer vision) Quantitative Imaging Methods and Translational Developments–Honoring the Memory of Dr. Larry Clarke 030220 oncology & carcinogenesis Medicine Radiology Nuclear Medicine and imaging National Lung Screening Trial Artificial intelligence business |
Zdroj: | Journal of medical imaging (Bellingham, Wash.). 5(1) |
ISSN: | 2329-4302 |
Popis: | 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. |
Databáze: | OpenAIRE |
Externí odkaz: |