Four-dimensional magnetic resonance imaging (4D-MRI) is an emerging technique for tumor motion management in image-guided radiation therapy (IGRT). However, current 4D-MRI suffers from low spatial resolution and strong motion artifacts owing to the long acquisition time and patients’ respiratory variations. If not managed properly, these limitations can adversely affect treatment planning and delivery in IGRT. In this study, we developed a novel deep learning framework called the coarse–super-resolution–fine network (CoSF-Net) to achieve simultaneous motion estimation and super-resolution within a unified model. We designed CoSF-Net by fully excavating the inherent properties of 4D-MRI, with consideration of limited and imperfectly matched training datasets. We conducted extensive experiments on multiple real patient datasets to assess the feasibility and robustness of the developed network. Compared with existing networks and three state-of-the-art conventional algorithms, CoSF-Net not only accurately estimated the deformable vector fields between the respiratory phases of 4D-MRI but also simultaneously improved the spatial resolution of 4D-MRI, enhancing anatomical features and producing 4D-MR images with high spatiotemporal resolution.

Feature selection has a direct impact on text categorization. Most existing algorithms are based on document level, and they haven’t considered the influence of term frequency on text categorization. Based on these, we put forward a feature selection approach, FSATD, based on term distributions in the paper. In our proposed algorithm, three critical factors which are term frequency, the inter-class distribution and the intra-class distribution of the terms are all considered synthetically. Finally, experiments are made with the help of kNN classifier. And the corresponding results on 20NewsGroup and SougouCS corpus show that FSATD algorithm achieves better performance than DF and t-Test algorithms.

Recently, deep learning has been demonstrated to be feasible in eliminating the use of gadolinium-based contrast agents (GBCAs) through synthesizing gadolinium-free contrast-enhanced MRI (GFCE-MRI) from contrast-free MRI sequences, pro-viding the community with an alternative to get rid of GBCAs-associated safety issues in patients. Nevertheless, generalizability assessment of the GFCE-MRI model has been largely challenged by the high inter-institutional heterogeneity of MRI data, on top of the scarcity of multi-institutional data itself. Although various data normalization methods have been adopted in previous studies to address the heterogeneity issue, it has been limited to single-institutional investigation and there is no standard normalization approach presently. In this study, we aimed at investigating gener-alizability of GFCE-MRI model using data from seven institutions by manipulating heterogeneity of training MRI data under five popular normalization approaches. Three state-of-the-art neural networks were applied to map from T1-weighted and T2-weighted MRI to contrast-enhanced MRI (CE-MRI) for GFCE-MRI synthesis in patients with nasopharyngeal carcinoma. MRI data from three institutions were used separately to generate three uni-institution models and jointly for a tri-institution model. The five normalization methods were applied to normalize the training and testing data of each model. MRI data from the remaining four institutions served as external cohorts for model generalizability assessment. Quality of GFCE-MRI was quantitatively evaluated against ground-truth CE-MRI using mean absolute error (MAE) and peak signal-to-noise ratio (PSNR). Results showed that performance of all uni-institution models remarkably dropped on the external cohorts. By contrast, model trained using multi-institutional data with Z-Score normalization yielded the best model generalizability improvement.

The key of swaying tree simulation is to capture the essence of the interaction between wind and tree, which has not received much attention from scholars. In this paper, we present a method for simulating swaying tree based on slicing partition. The tree is divided into multiple “slices”, and then jointly moved to simulate the tree swaying in wind. By slicing partition, the motion of any slice is independent of the others concurrently, which provides a parallel spatial topology for real-time calculation. Moreover, based on the fact that the wind will blow to a tree partly rather than the whole, partial swayings of the tree are also simulated. The experimental results show the correctness and effectiveness of our method.

During the data acquisition procedure, volume data are usually contaminated by noises. This would create visual confusion and misunderstanding in analyzing the volume data. Thus noise reduction is necessary for improving the quality of volume data inspection and analytic tasks. However, it is far from being fully resolved for removing noise while maximally retaining geometric sharp features. In the work, we present a powerful volume denoising method based on the extended weighted least squares. We improve the weighted least squares method and extend it to 3D for volume data denoising. The primary advantage of the proposed method is that it can consistently produce better results for removing noise while preserving sharp features. We illustrate our technique on synthetic and real-world 3D data and compare our method with median method, weighted least squares, L0 volume gradient minimization and edge aware anisotropic diffusion method, the experimental results demonstrate that our method can achieve higher quality results than the selected state-of-the art methods.