Dataset Information

Toward automatic prediction of EGFR mutation status in pulmonary adenocarcinoma with 3D deep learning.

ABSTRACT: To develop a deep learning system based on 3D convolutional neural networks (CNNs), and to automatically predict EGFR-mutant pulmonary adenocarcinoma in CT images. A dataset of 579 nodules with EGFR mutation status labels of mutant (Mut) or wild-type (WT) was retrospectively analyzed. A deep learning system, namely 3D DenseNets, was developed to process 3D patches of nodules from CT data, and learn strong representations with supervised end-to-end training. The 3D DenseNets were trained with a training subset of 348 nodules and tuned with a development subset of 116 nodules. A strong data augmentation technique, mixup, was used for better generalization. We evaluated our model on a holdout subset of 115 nodules. An independent public dataset of 37 nodules from the cancer imaging archive (TCIA) was also used to test the generalization of our method. Conventional radiomics analysis was also performed for comparison. Our method achieved promising performance on predicting EGFR mutation status, with AUCs of 75.8% and 75.0% for our holdout test set and public test set, respectively. Moreover, strong relations were found between deep learning feature and conventional radiomics, while deep learning worked through an enhanced radiomics manner, that is, deep learned radiomics (DLR), in terms of robustness, compactness and expressiveness. The proposed deep learning system predicts EGFR-mutant of lung adenocarcinomas in CT images noninvasively and automatically, indicating its potential to help clinical decision-making by identifying eligible patients of pulmonary adenocarcinoma for EGFR-targeted therapy.

SUBMITTER: Zhao W

PROVIDER: S-EPMC6601587 | biostudies-literature | 2019 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Toward automatic prediction of EGFR mutation status in pulmonary adenocarcinoma with 3D deep learning.

Zhao Wei W Yang Jiancheng J Ni Bingbing B Bi Dexi D Sun Yingli Y Xu Mengdi M Zhu Xiaoxia X Li Cheng C Jin Liang L Gao Pan P Wang Peijun P Hua Yanqing Y Li Ming M

Cancer medicine 20190510 7

To develop a deep learning system based on 3D convolutional neural networks (CNNs), and to automatically predict EGFR-mutant pulmonary adenocarcinoma in CT images. A dataset of 579 nodules with EGFR mutation status labels of mutant (Mut) or wild-type (WT) was retrospectively analyzed. A deep learning system, namely 3D DenseNets, was developed to process 3D patches of nodules from CT data, and learn strong representations with supervised end-to-end training. The 3D DenseNets were trained with a t ...[more]

PMID: 31074592

Similar Datasets

Project description:To recognize the epidermal growth factor receptor (EGFR) gene mutation status in lung adenocarcinoma (LADC) has become a prerequisite of deciding whether EGFR-tyrosine kinase inhibitor (EGFR-TKI) medicine can be used. Polymerase chain reaction assay or gene sequencing is for measuring EGFR status, however, the tissue samples by surgery or biopsy are required. We propose to develop deep learning models to recognize EGFR status by using radiomics features extracted from non-invasive CT images. Preoperative CT images, EGFR mutation status and clinical data have been collected in a cohort of 709 patients (the primary cohort) and an independent cohort of 205 patients. After 1,037 CT-based radiomics features are extracted from each lesion region, 784 discriminative features are selected for analysis and construct a feature mapping. One Squeeze-and-Excitation (SE) Convolutional Neural Network (SE-CNN) has been designed and trained to recognize EGFR status from the radiomics feature mapping. SE-CNN model is trained and validated by using 638 patients from the primary cohort, tested by using the rest 71 patients (the internal test cohort), and further tested by using the independent 205 patients (the external test cohort). Furthermore, SE-CNN model is compared with machine learning (ML) models using radiomics features, clinical features, and both features. EGFR(-) patients show the smaller age, higher odds of female, larger lesion volumes, and lower odds of subtype of acinar predominant adenocarcinoma (APA), compared with EGFR(+). The most discriminative features are for texture (614, 78.3%) and the features of first order of intensity (158, 20.1%) and the shape features (12, 1.5%) follow. SE-CNN model can recognize EGFR mutation status with an AUC of 0.910 and 0.841 for the internal and external test cohorts, respectively. It outperforms the CNN model without SE, the fine-tuned VGG16 and VGG19, three ML models, and the state-of-art models. Utilizing radiomics feature mapping extracted from non-invasive CT images, SE-CNN can precisely recognize EGFR mutation status of LADC patients. The proposed method combining radiomics features and deep leaning is superior to ML methods and can be expanded to other medical applications. The proposed SE-CNN model may help make decision on usage of EGFR-TKI medicine.

Project description:IntroductionIn lung adenocarcinoma, the mutational spectrum is dominated by EGFR and KRAS mutations. Improved knowledge about genomic and transcriptional alterations in and between mutation-defined subgroups may identify genes involved in disease development or progression.MethodsGenomic profiles from 457 adenocarcinomas, including 113 EGFR-mutated, 134 KRAS-mutated and 210 EGFR and KRAS-wild type tumors (EGFRwt/KRASwt), and gene expression profiles from 914 adenocarcinomas, including 309 EGFR-mutated, 192 KRAS-mutated, and 413 EGFRwt/KRASwt tumors, were assembled from different repositories. Genomic and transcriptional differences between the three mutational groups were analyzed by both supervised and unsupervised methods.ResultsEGFR-mutated adenocarcinomas displayed a larger number of copy number alterations and recurrent amplifications, a higher fraction of total loss-of-heterozygosity, higher genomic complexity, and a more distinct expression pattern than EGFR-wild type adenocarcinomas. Several of these differences were also consistent when the three mutational groups were stratified by stage, gender and smoking status. Specific copy number alterations were associated with mutation status, predominantly including regions of gain with the highest frequency in EGFR-mutated tumors. Differential regions included both large and small regions of gain on 1p, 5q34-q35.3, 7p, 7q11.21, 12p12.1, 16p, and 21q, and losses on 6q16.3-q21, 8p, and 9p, with 20-40% frequency differences between the mutational groups. Supervised gene expression analyses identified 96 consistently differentially expressed genes between the mutational groups, and together with unsupervised analyses these analyses highlighted the difficulty in broadly resolving the three mutational groups into distinct transcriptional entities.ConclusionsWe provide a comprehensive overview of the genomic and transcriptional landscape in lung adenocarcinoma stratified by EGFR and KRAS mutations. Our analyses suggest that the overall genomic and transcriptional landscape of lung adenocarcinoma is affected, but only to a minor extent, by EGFR and KRAS mutation status.

Dataset Information

Toward automatic prediction of EGFR mutation status in pulmonary adenocarcinoma with 3D deep learning.

Publications

Toward automatic prediction of EGFR mutation status in pulmonary adenocarcinoma with 3D deep learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets