Abstract
Deep learning-based radiomics (DLR) was developed to extract deep information from multiple modalities of magnetic resonance (MR) images. The performance of DLR for predicting the mutation status of isocitrate dehydrogenase 1 (IDH1) was validated in a dataset of 151 patients with low-grade glioma. A modified convolutional neural network (CNN) structure with 6 convolutional layers and a fully connected layer with 4096 neurons was used to segment tumors. Instead of calculating image features from segmented images, as typically performed for normal radiomics approaches, image features were obtained by normalizing the information of the last convolutional layers of the CNN. Fisher vector was used to encode the CNN features from image slices of different sizes. High-throughput features with dimensionality greater than 1.6*104 were obtained from the CNN. Paired t-tests and F-scores were used to select CNN features that were able to discriminate IDH1. With the same dataset, the area under the operating characteristic curve (AUC) of the normal radiomics method was 86% for IDH1 estimation, whereas for DLR the AUC was 92%. The AUC of IDH1 estimation was further improved to 95% using DLR based on multiple-modality MR images. DLR could be a powerful way to extract deep information from medical images.
Highlights
Of necrosis, enhancements and edema regions of interests (ROIs)
convolutional neural network (CNN) is a representative method used for deep learning, and it has been successfully applied to the field of image segmentation[8]
In glioma segmentation based on magnetic resonance (MR) images, most of the CNN methods were proposed for high-grade gliomas[10, 11]
Summary
Of necrosis, enhancements and edema regions of interests (ROIs). Their results showed that a certain image features correlated with molecular subgroups. But not least, current radiomics methods often characterize medical images by using several groups of image features, including intensity, shape, texture and wavelets. Many such image features can be calculated, it is not possible for all these imaging characteristics of segmented areas to be included in the predesigned features. DLR obtains radiomics features by normalizing the information from a deep neural network designed for image segmentation. In DLR, the high-throughput image features are directly extracted from the deep neural network. Because DLR does not involve extra feature extraction operations, no extra errors will be introduced into the radiomics analysis because of feature calculations. Many groups have used CNN for the segmentation of medical images, and it has provided better results than traditional methods[9]. A Fisher vector is used to normalize the network information from MR imaging slices of different sizes; 16,384 high-throughput image features were generated from the CNN for each case
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
