Abstract
The mutation status of the epidermal growth factor receptor (EGFR) is an important clinical reference indicator for lung cancer diagnosis and treatment. However, the extraction of effective discriminative features for non-invasive computer-aided EGFR mutation prediction still poses a big challenge. In this paper, multiple types of features are designed and analyzed to address this problem. These features include clinical features based on prior medical knowledge and quantitative image features extracted by convolutional neural networks (CNN). A long short-term memory (LSTM) network is also introduced to exploit the dependency between these feature types and then fuse them. In particular, a CNN is constructed to extract quantitative features of computed-tomography (CT) images. Furthermore, a LSTM is utilized to analyze the dependency between these clinical and CT image features and generate a new feature representation for computer-aided diagnosis. For samples from the same category, the proposed method deal with feature representation variabilities arising from interdependencies in multi-type features and patient specificity. The multiple feature types of the collected clinical data are used to assess the proposed approach and other relevant algorithms. Our results demonstrate that the multi-type dependency-based feature representation shows superior performance (Accuracy = 75%, AUC = 0.78) compared to single-type feature representations. The proposed method is reliable to apply for diagnosing of the EGFR mutation status.
Highlights
Cancer has become the greatest threat to human health since the beginning of the 21st century
Patients who met the following criteria were included in this dataset: (1) Having an epidermal growth factor receptor (EGFR) mutation type of lung adenocarcinoma with gold standard verification; (2) availability of CT imaging data; (3) availability of pathology reports with verified diagnosis by experienced radiologists; (4) availability of clinical data, including age, sex as well as smoking history
convolutional neural network (CNN) is used to extract image quantitative feature while long short-term memory (LSTM) is combined to model the relationship between different types of features, achieving feature fusion
Summary
Cancer has become the greatest threat to human health since the beginning of the 21st century. Exploring effective techniques for the rapid diagnosis and treatment of lung cancer has recently become a hot topic of research. Many investigations considered a specific related genetic phenotype, namely, the epidermal growth factor receptor (EGFR) which has become a standard for the treatment of lung adenocarcinoma [2], [3]. The gold standard for EGFR is mutational sequencing of biopsies. Due to the heterogeneity of lung tumors, EGFR-based methods require repeated sampling of tumor tissues, which causes inconvenience, costly treatment, and poor detection efficiency [7], [8]. As the first step in clinical diagnosis, a noninvasive approach serves as a useful aid in the early planning for cost-effective treatment. A non-invasive computeraided diagnosis method has become increasingly prominent as a viable alternative for reducing the inherent risk of the biopsy surgical procedures [6]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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.
