백종덕 교수

PURPOSE: Convolutional neural network (CNN)-based image denoising techniques have shown promising results in low-dose CT denoising. However, CNN often introduces blurring in denoised images when trained with a widely used pixel-level loss function. Perceptual loss and adversarial loss have been proposed recently to further improve the image denoising performance. In this paper, we investigate the effect of different loss functions on image denoising performance using task-based image quality ass

The noise power spectrum (NPS) is a useful metric for understanding the noise content in images. To examine some unique properties of the NPS of fan beam CT, the authors derived an analytical expression for the NPS of fan beam CT and validated it with computer simulations. The nonstationary noise behavior of fan beam CT was examined by analyzing local regions and the entire field-of-view (FOV). This was performed for cases with uniform as well as nonuniform noise across the detector cells and ac

We examine the noise advantages of having a computed tomography (CT) detector whose spatial resolution is significantly better (e.g. a factor of 2) than needed for a desired resolution in the reconstructed images. The effective resolution of detectors in x-ray CT is sometimes degraded by binning cells because the small cell size and fine sampling are not needed to achieve the desired resolution (e.g. with flat panel detectors). We studied the effect of the binning process on the noise in the rec

PURPOSE: The authors examine the nonstationary noise behavior of a cone-beam CT system with FDK reconstruction. METHODS: To investigate the nonstationary noise behavior, an analytical expression for the NPS of local volumes and an entire volume was derived and quantitatively compared to the NPS estimated from experimental air and water images. RESULTS: The NPS of local volumes at different locations along the z-axis showed radial symmetry in the f(x)-f(y) plane and different missing cone regions

PURPOSE: Convolutional neural network (CNN)-based denoising is an effective method for reducing complex computed tomography (CT) noise. However, the image blur induced by denoising processes is a major concern. The main source of image blur is the pixel-level loss (e.g., mean squared error [MSE] and mean absolute error [MAE]) used to train a CNN denoiser. To reduce the image blur, feature-level loss is utilized to train a CNN denoiser. A CNN denoiser trained using visual geometry group (VGG) los

PURPOSE: Sparse-view sampling has attracted attention for reducing the scan time and radiation dose of dental cone-beam computed tomography (CBCT). Recently, various deep learning-based image reconstruction techniques for sparse-view CT have been employed to produce high-quality image while effectively reducing streak artifacts caused by the lack of projection views. However, most of these methods do not fully consider the effects of metal implants. As sparse-view sampling strengthens the artifa

PURPOSE: In this paper, we propose a convolutional neural network (CNN)-based efficient model observer for breast computed tomography (CT) images. METHODS: We first showed that the CNN-based model observer provided similar detection performance to the ideal observer (IO) for signal-known-exactly and background-known-exactly detection tasks with an uncorrelated Gaussian background noise image. We then demonstrated that a single-layer CNN without a nonlinear activation function provided similar de

We investigate the detectability of breast cone beam computed tomography images using human and model observers and the variations of exponent, β, of the inverse power-law spectrum for various reconstruction filters and interpolation methods in the Feldkamp-Davis-Kress (FDK) reconstruction. Using computer simulation, a breast volume with a 50% volume glandular fraction and a 2mm diameter lesion are generated and projection data are acquired. In the FDK reconstruction, projection data are apodize

PURPOSE: Sparse-view computed tomography (CT) has been attracting attention for its reduced radiation dose and scanning time. However, analytical image reconstruction methods suffer from streak artifacts due to insufficient projection views. Recently, various deep learning-based methods have been developed to solve this ill-posed inverse problem. Despite their promising results, they are easily overfitted to the training data, showing limited generalizability to unseen systems and patients. In t

We present initial experimental results of a rotating-gantry multi-source inverse-geometry CT (MS-IGCT) system. The MS-IGCT system was built with a single module of 2 × 4 x-ray sources and a 2D detector array. It produced a 75 mm in-plane field-of-view (FOV) with 160 mm axial coverage in a single gantry rotation. To evaluate system performance, a 2.5 inch diameter uniform PMMA cylinder phantom, a 200 µm diameter tungsten wire, and a euthanized rat were scanned. Each scan acquired 125 views per s

The purpose of this study is implementation of an anthropomorphic model observer using a convolutional neural network (CNN) for signal-known-statistically (SKS) and background-known-statistically (BKS) detection tasks. We conduct SKS/BKS detection tasks on simulated cone beam computed tomography (CBCT) images with eight types of signal and randomly varied breast anatomical backgrounds. To predict human observer performance, we use conventional anthropomorphic model observers (i.e. the non-prewhi

The shape of the NPS depends on the detector lag coefficients, location of the region, and the number of views used in the reconstruction. In general, the noise correlation caused by detector lag decreased the amplitude of the NPS.

To assess the resolution performance of modern CT scanners, a method to measure the 3D MTF is needed. Computationally, a point object is an ideal test phantom but is difficult to apply experimentally. Recently, Thornton et al. described a method to measure the directional MTF using a sphere phantom. We tested this method for FDK reconstructions by simulating a sphere and a point object centered at (0.01 cm , 0.01 cm, 0.01 cm) and (0.01 cm, 0.01 cm, 10.01 cm) and compared the directional MTF esti