In this work, a novel cascaded neural network architecture was developed to perform cone beam CT image reconstruction using the deep learning method. The proposed architecture consists four individual stages: a manifold learning stage to perform projection data pre-processing, a convolutional neural network (CNN) stage to perform data filtration, a fully connected layer with sparse regularization to perform single-view backprojection, and a final fully connected layer with linear activation to generate the target image volume. In manifold learning stage, a novel feature combining technique was proposed to improve noise properties of the final reconstructed images. These 13-layer deep neural network work trained using extensive numerical phantom with noise contaminated projection data and ground truth image in a stage-by-stage pretraining stage. After pretraining with numerical phantom data, the cascaded neural network model was fine tuned using physical phantom data from a diagnostic MDCT scanner. After training, the trained neural network model was used to reconstruct low dose CT images for human subjects from a prospective low dose CT protocol. In these studies, it was found that the proposed cascaded neural network based deep learning method can (1) enable low dose CT reconstruction without noise streaks and with reduced noise amplitude; (2) well maintain reconstruction accuracy at reduced dose levels; and (3) unlike the currently available statistical model based image reconstruction (MBIR) methods, the proposed deep learning reconstruction method can well maintain the similar dose-normalized noise power spectrum (NPS) with that of the FBP reconstructed images.
Yinsheng Li and Guang-Hong Chen, "Deep learning based cone beam CT reconstruction framework using a cascaded neural network architecture (Conference Presentation)," Proc. SPIE 10573, Medical Imaging 2018: Physics of Medical Imaging, 105731J (Presented at SPIE Medical Imaging: February 14, 2018; Published: 14 March 2018); https://doi.org/10.1117/12.2293916.5750740887001.
Conference Presentations are recordings of oral presentations given at SPIE conferences and published as part of the conference proceedings. They include the speaker's narration along with a video recording of the presentation slides and animations. Many conference presentations also include full-text papers. Search and browse our growing collection of more than 14,000 conference presentations, including many plenary and keynote presentations.
Study of self-shadowing effect as a simple means to realize nanostructured thin films and layers with special attentions to birefringent obliquely deposited thin films and photo-luminescent porous silicon