Noise reduction with low dose CT data based on a modified ROF model.

In order to reduce the radiation exposure caused by Computed Tomography (CT) scanning, low dose CT has gained much interest in research as well as in industry. One fundamental difficulty for low dose CT lies in its heavy noise pollution in the raw data which leads to quality deterioration for reconstructed images. In this paper, we propose a modified ROF model to denoise low dose CT measurement data in light of Poisson noise model. Experimental results indicate that the reconstructed CT images based on measurement data processed by our model are in better quality, compared to the original ROF model or bilateral filtering.

Anti-hyperlipidemic and ameliorative effects of chickpea starch and resistant starch in mice with high fat diet induced obesity are associated with their multi-scale structural characteristics.

Chickpea starches were isolated from both untreated (UC-S) and conventionally cooked seeds (CC-S), and their multi-scale structural characteristics and in vivo physiological effects on controlling hyperlipidemia in high fat diet induced obese mice were compared with their corresponding resistant starch (RS) fractions obtained by an in vitro enzymatic isolation method (UC-RS and CC-RS). The degree of order/degree of double helix in Fourier transform infrared spectroscopy was in the following order: CC-RS > UC-RS > CC-S > UC-S, which was consistent with the trend observed for relative crystallinity and double helix contents monitored by X-ray diffractometer and solid-state 13C cross-polarization and magic angle spinning NMR analyses. The influence of different types of chickpea starch and their corresponding resistant starch fractions on regulating the serum lipid profile, antioxidant status, and histopathological changes in liver, colon and cecal tissues, and gene expressions associated with lipid metabolism, gut microbiota, as well as short-chain fatty acid metabolites in mice with high fat diet induced obesity was investigated. The results showed that the chickpea RS diet group exhibited overall better anti-hyperlipidemic and ameliorative effects than those of the starch group, and such effects were most pronounced in the CC-RS intervention group. After a six-week period of administration with chickpea starch and RS diets, mice in the UC-RS and CC-RS groups tended to have relatively significantly higher levels (P < 0.05) of butyric acid in their fecal contents. The 16S rRNA sequencing results revealed that mice fed with CC-RS showed the greatest abundance of Akkermansia and Lactobacillus compared with the other groups.

Skin3D: Detection and longitudinal tracking of pigmented skin lesions in 3D total-body textured meshes.

We present an automated approach to detect and longitudinally track skin lesions on 3D total-body skin surface scans. The acquired 3D mesh of the subject is unwrapped to a 2D texture image, where a trained objected detection model, Faster R-CNN, localizes the lesions within the 2D domain. These detected skin lesions are mapped back to the 3D surface of the subject and, for subjects imaged multiple times, we construct a graph-based matching procedure to longitudinally track lesions that considers the anatomical correspondences among pairs of meshes and the geodesic proximity of corresponding lesions and the inter-lesion geodesic distances. We evaluated the proposed approach using 3DBodyTex, a publicly available dataset composed of 3D scans imaging the coloured skin (textured meshes) of 200 human subjects. We manually annotated locations that appeared to the human eye to contain a pigmented skin lesion as well as tracked a subset of lesions occurring on the same subject imaged in different poses. Our results, when compared to three human annotators, suggest that the trained Faster R-CNN detects lesions at a similar performance level as the human annotators. Our lesion tracking algorithm achieves an average matching accuracy of 88% on a set of detected corresponding pairs of prominent lesions of subjects imaged in different poses, and an average longitudinal accuracy of 71% when encompassing additional errors due to lesion detection. As there currently is no other large-scale publicly available dataset of 3D total-body skin lesions, we publicly release over 25,000 3DBodyTex manual annotations, which we hope will further research on total-body skin lesion analysis.

Micro-CT artifacts reduction based on detector random shifting and fast data inpainting.

PURPOSE: In Micro-CT systems based on optical coupling detectors, the defects of scintillator or CCD-camera would lead to heavy artifacts in reconstructed CT images. Meanwhile, different detector units usually suffer from inhomogeneous response, which also leads to artifacts in the CT images. Detector shifting is a simple and efficient method to remove the artifacts due to inhomogeneous responses of detector units. However, it does not work well for heavy artifacts due to defects in scintillator or CCD. The purpose of this paper is to develop a data preprocessing method to reduce both kinds of artifacts. METHODS: A hybrid method which involves detector random shifting and data inpainting is proposed to correct the projection data, so as to suppress the artifacts in the reconstructed CT images. The defects in scintillator or CCD-camera lead to data lost in some areas of the projection data. The Criminisi algorithm is employed to recover the lost data. By detector random shifting, the location of the lost data in one view might be shifted away in adjacent views. This feature is utilized to design the search window, such that the best match patch shall be searched across adjacent views. By this way, the best match patches should really enjoy high similarity. As a result, the heavy artifacts due to defects of scintillator or CCD-camera should be suppressed. Furthermore, a multiscale tessellation method is proposed to locate the defects and similarity patches, which makes the Criminisi algorithm very fast. RESULTS: The authors tested the proposed method on both simulated projection data and real projection data. Experiments show that the proposed method could correct the bad data in the projections quite well. Compared to other popular methods, such as linear interpolation, wavelet combining Fourier transform, and TV-inpainting, experimental results suggest that the CT images reconstructed from the preprocessed data sets by our method is significantly better in quality. CONCLUSIONS: They have proposed a hybrid method for projection data preprocessing which fits well to typical Micro-CT systems. The hybrid method could suppress the ring artifacts in the reconstructed CT images efficiently, while the spatial resolution is not reduced even with a critical eye.