Analyzing the geometric and semantic properties of 3D point clouds through deep networks is still challenging due to irregularity sparsity samplings their structures. This paper presents a new method define compute convolution directly on by proposed annular convolution. operator can better capture local neighborhood geometry each specifying (regular dilated) ring-shaped structures directions in computation. It adapt variability scalability at signal processing level. We apply it developed...

Better understanding of the dose-toxicity relationship is critical for safe dose escalation to improve local control in late-stage cervical cancer radiotherapy. In this study, we introduced a convolutional neural network (CNN) model analyze rectum distribution and predict toxicity. Forty-two patients treated with combined external beam radiotherapy (EBRT) brachytherapy (BT) were retrospectively collected, including twelve toxicity thirty non-toxicity patients. We adopted transfer learning...

The anisotropy of mechanical strength on a 3D printed model can be controlled in multi-axis printing system as materials accumulated along dynamically varied directions. In this paper, we present new computational framework to generate specially designed layers and toolpaths for strengthening by aligning filaments the directions with large stresses. major challenge comes from how effectively decompose solid into sequence strength-aware collision-free working surfaces. We formulate it problem...

This paper introduces a particle-based approach for anisotropic surface meshing. Given an input polygonal mesh endowed with Riemannian metric and specified number of vertices, the method generates metric-adapted mesh. The main idea consists mapping space into higher dimensional isotropic one, called "embedding space". vertices are generated by uniformly sampling in this embedding space, is further regularized optimizing energy function quasi-Newton algorithm. All computations can be...

This paper introduces a deep neural network based method, i.e., DeepOrganNet, to generate and visualize fully high-fidelity 3D / 4D organ geometric models from single-view medical images with complicated background in real time. Traditional image reconstruction requires near hundreds of projections, which cost insufferable computational time deliver undesirable high imaging radiation dose human subjects. Moreover, it always needs further notorious processes segment or extract the accurate...

Previous approaches on 3D shape segmentation mostly rely heuristic processing and hand-tuned geometric descriptors. In this paper, we propose a novel representation learning approach, Directionally Convolutional Network (DCN), to solve the problem. DCN extends convolution operations from images surface mesh of shapes. With DCN, learn effective representations raw features, i.e., face normals distances, achieve robust segmentation. More specifically, two-stream framework is proposed: one...

The fundamental motivation of the proposed work is to present a new visualization-guided computing paradigm combine direct 3D volume processing and rendered clues for effective exploration. For example, extracting visualizing microstructures in-vivo have been long-standing challenging problem. However, due high sparseness noisiness in cerebrovasculature data as well highly complex geometry topology variations micro vessels, it still extremely extract complete vessel structure visualize with...

Reducing nurse job burnout is vital for quality care and turnover reduction, particularly in emergency departments. Given that moral distress a crucial predictor of burnout, this study seeks to identify factors can alter relationship its underlying mechanisms. The finding essential enhancing satisfaction among nurses improving patient safety healthcare quality. This employed cross-sectional design was conducted May 2024 the departments five tertiary hospitals Southern China. survey...

Better knowledge of the dose-toxicity relationship is essential for safe dose escalation to improve local control in cervical cancer radiotherapy. The conventional model based on volume histogram, which parameter lacking spatial information. To overcome this limit, we explore a comprehensive rectal both histogram and map features accurate radiation toxicity prediction. Forty-two patients treated with combined external beam radiotherapy (EBRT) brachytherapy (BT) were retrospectively studied,...

Abstract Sparse localized decomposition is a useful technique to extract meaningful deformation components out of training set mesh data. However, existing methods cannot capture large rotational motion in the given dataset. In this paper we present new based on gradients. Given dataset, gradient field extracted, and decomposed into two groups: rotation stretching field, through polar decomposition. These groups information are further processed sparse desired components. can be linearly...

In mesh simplification, common requirements like accuracy, triangle quality, and feature alignment are often considered as a trade-off. Existing algorithms concentrate on just one or few specific aspects of these requirements. For example, the well-known Quadric Error Metrics (QEM) approach [Garland Heckbert 1997] prioritizes accuracy can preserve strong lines/points well, but falls short in ensuring high quality may degrade weak features that not distinctive ones. this paper, we propose...

Analyzing the geometric and semantic properties of 3D point clouds through deep networks is still challenging due to irregularity sparsity samplings their structures. This paper presents a new method define compute convolution directly on by proposed annular convolution. operator can better capture local neighborhood geometry each specifying (regular dilated) ring-shaped structures directions in computation. It adapt variability scalability at signal processing level. We apply it developed...

Vector image representation methods that can faithfully reconstruct objects and color variations in a raster are desired many practical applications. This article presents triangular configuration B-spline (referred to as TCB-spline)-based vector graphics for vectorization. Based on this new representation, an automatic vectorization paradigm is proposed. The proposed framework first detects sharp curvilinear features the constructs knot meshes based detected feature lines. It iteratively...

A simultaneous motion estimation and image reconstruction (SMEIR) strategy was proposed for 4D cone-beam CT (4D-CBCT) showed excellent results in both phantom lung cancer patient studies. In the original SMEIR algorithm, deformation vector field (DVF) defined on voxel grid estimated by enforcing a global smoothness regularization term fields. The objective of this work is to improve computation efficiency accuracy 4D-CBCT through developing multi-organ meshing model. Feature-based adaptive...

Volatile organic compounds (VOCs) could reflect changes resulting from ongoing pathophysiological processes and altered body metabolisms, thus have been studied for various types of cancers. We aimed to test an advanced global metabolomic technique characterize circulating VOCs in patients diagnosed with colorectal cancer (CRC). employed solid-phase microextraction (SPME) comprehensive two-dimensional gas chromatography mass-spectrometry (GC × GC–MS). analyzed 30 random plasma samples...

The aim of this study is to develop an internal-external correlation model for internal motion estimation lung cancer radiotherapy. Deformation vector fields that characterize the are obtained by respectively registering organ meshes and external surface from 4DCT images via a recently developed local topology preserved non-rigid point matching algorithm. A composite matrix constructed combing estimated phasic DVFs with directional DVFs. Principle component analysis then applied extract...