Automatic Lung Segmentation Based on Texture and Deep Features of HRCT Images with Interstitial Lung Disease

• Published in: BioMed research international 2019, Vol.2019 (2019), p.1-8
• Main Authors: Zhao, Lijie, Zhang, Xinwang, Guo, Shaoyong, Pang, Ting
• Format: Article
• Language: English
• Subjects: Algorithms; Analysis; Artificial neural networks; Computed tomography; CT imaging; Databases, Factual; Deep learning; Diagnosis, Computer-Assisted - methods; Diagnostic imaging; Feature extraction; Female; Fuzzy sets; Humans; Image processing; Image resolution; Image segmentation; International conferences; Lung - diagnostic imaging; Lung - pathology; Lung diseases; Lung Diseases, Interstitial - diagnostic imaging; Male; Mammography; Medical imaging; Medical imaging equipment; Methods; Middle Aged; Models, Theoretical; Neural networks; Neural Networks, Computer; Noise; Noise reduction; Pattern recognition; Radiomics; Signal processing; Texture; Tomography, X-Ray Computed - methods; Training; Wavelet transforms; Wiener filtering
• ISSN: 2314-6133; 2314-6141
• DOI: 10.1155/2019/2045432

Lung segmentation in high-resolution computed tomography (HRCT) images is necessary before the computer-aided diagnosis (CAD) of interstitial lung disease (ILD). Traditional methods are less intelligent and have lower accuracy of segmentation. This paper develops a novel automatic segmentation model using radiomics with a combination of hand-crafted features and deep features. The study uses ILD Database-MedGIFT from 128 patients with 108 annotated image series and selects 1946 regions of interest (ROI) of lung tissue patterns for training and testing. First, images are denoised by Wiener filter. Then, segmentation is performed by fusion of features that are extracted from the gray-level co-occurrence matrix (GLCM) which is a classic texture analysis method and U-Net which is a standard convolutional neural network (CNN). The final experiment result for segmentation in terms of dice similarity coefficient (DSC) is 89.42%, which is comparable to the state-of-the-art methods. The training performance shows the effectiveness for a combination of texture and deep radiomics features in lung segmentation.