Date of Submission

Fall 12-17-2019

Degree Type

Thesis

Degree Name

Master of Science in Computer Science (MSCS)

Department

Computer Science

Track

Big Data

Chair

Dr. Jing (Selena) He

Committee Member

Dr. Meng Han

Committee Member

Dr. Xiaohua Xu

Abstract

Interstitial lung disease (ILD) causes pulmonary fibrosis. The correct classification of ILD plays a crucial role in the diagnosis and treatment process. In this research work, we disclose a lung nodules recognition method based on a deep convolutional neural network (DCNN) and global features, which can be used for computer-aided diagnosis (CAD) of global features of lung nodules. Firstly, a DCNN is constructed based on the characteristics and complexity of lung computerized tomography (CT) images. Then discussed the effects of different iterations on the recognition results and influence of different model structures on the global features of lung nodules. We also improved the convolution kernel size, feature dimension, and network depth. Finally, the effects of different pooling methods, activation functions and training algorithms on the performance of DCNN were analyzed from the network optimization dimension. The experimental results verify the feasibility of the proposed DCNN for CAD of global features of lung nodules. Selecting appropriate model parameters and model structure and using the elastic momentum training method can achieve good recognition results.

Share

COinS