Search

Your search keyword '"Liu,Xudong"' showing total 3,683 results
Start Over Author "Liu,Xudong"
3,683 results on '"Liu,Xudong"'

213. Prognostic Model Based on Preoperative FAR and SII Versus TNM Staging System in Evaluating Prognosis of Patients with Pancreatic Cancer After Radical Resection

Author

LIU Xudong, ZHAO Bin, DU Peng, ZHANG Guoqiang, ZHENG Qiang, LAI Jiamin, and CHENG Zhibin

Subjects
pancreatic ductal adenocarcinoma, fibrinogen/albumin ratio, systemic immune inflammation index, nomogram, prognostic model, Neoplasms. Tumors. Oncology. Including cancer and carcinogens, RC254-282
Abstract

Objective To investigate the predictive value of preoperative fibrinogen/albumin ratio (FAR) and systemic immune inflammation index (SII) on the postoperative prognosis of patients with pancreatic ductal adenocarcinoma. Methods An ROC curve was used in determining the best cutoff values of FAR and SII and then grouped. The Cox proportional hazards model was used in analyzing the prognostic factors of radical pancreatic cancer surgery, and then a Nomogram prognostic model was established. C-index, AUC, and calibration curve were used in evaluating the discrimination and calibration ability of the Nomogram. DCA curves were used in assessing the clinical validity of the Nomograms. Results The optimal cutoff values for preoperative FAR and SII were 0.095 and 532.945, respectively. FAR≥ 0.095, SII≥ 532.945, CA199≥ 450.9 U/ml, maximum tumor diameter≥ 4 cm, and the absence of postoperative chemotherapy were independent risk factors for the poor prognosis of pancreatic cancer (P

Published
2023
Full Text
View/download PDF

220. Factor Structure of Repetitive Behaviors across Autism Spectrum Disorder and Attention-Deficit/Hyperactivity Disorder

Author

Brierley, Noah J., McDonnell, Christina G., Parks, Kaitlyn M. A., Schulz, Samantha E., Dalal, Tyler C., Kelley, Elizabeth, Anagnostou, Evdokia, Nicolson, Robert, Georgiades, Stelios, Crosbie, Jennifer, Schachar, Russell, Liu, Xudong, and Stevenson, Ryan A.

Abstract

Restricted interests and repetitive behaviors (RRBs) are core symptoms of autism spectrum disorder (ASD), and commonly occur in attention-deficit/hyperactivity disorder (ADHD). Little is known about how RRBs manifest in ADHD. We quantified and compared factor structures of RRBs in children with ASD (n = 634) or ADHD (n = 448), and related factors to sex and IQ. A four-factor solution emerged, including Stereotypy, Self-Injury, Compulsions, and Ritualistic/Sameness. Factor structures were equivalent across diagnoses, though symptoms were more severe in ASD. IQ negatively correlated with Stereotypy, Self-Injury, and Compulsions in ASD, and negatively correlated with Compulsions and Ritualistic/Sameness behaviors in ADHD. In ASD only, females exhibited higher Self-Injury. Thus, patterns of RRBs are preserved across ASD and ADHD, but severity and relationship with IQ differed.

Published
2021
Full Text
View/download PDF

221. Understanding Beauty via Deep Facial Features

Author

Liu, Xudong, Li, Tao, Peng, Hao, Ouyang, Iris Chuoying, Kim, Taehwan, and Wang, Ruizhe

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

The concept of beauty has been debated by philosophers and psychologists for centuries, but most definitions are subjective and metaphysical, and deficit in accuracy, generality, and scalability. In this paper, we present a novel study on mining beauty semantics of facial attributes based on big data, with an attempt to objectively construct descriptions of beauty in a quantitative manner. We first deploy a deep convolutional neural network (CNN) to extract facial attributes, and then investigate correlations between these features and attractiveness on two large-scale datasets labelled with beauty scores. Not only do we discover the secrets of beauty verified by statistical significance tests, our findings also align perfectly with existing psychological studies that, e.g., small nose, high cheekbones, and femininity contribute to attractiveness. We further leverage these high-level representations to original images by a generative adversarial network (GAN). Beauty enhancements after synthesis are visually compelling and statistically convincing verified by a user survey of 10,000 data points.

Published
2019

232. Semi-supervised Text Regression with Conditional Generative Adversarial Networks

Author

Li, Tao, Liu, Xudong, and Su, Shihan

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence, Quantitative Finance - Computational Finance, Statistics - Machine Learning
Abstract

Enormous online textual information provides intriguing opportunities for understandings of social and economic semantics. In this paper, we propose a novel text regression model based on a conditional generative adversarial network (GAN), with an attempt to associate textual data and social outcomes in a semi-supervised manner. Besides promising potential of predicting capabilities, our superiorities are twofold: (i) the model works with unbalanced datasets of limited labelled data, which align with real-world scenarios; and (ii) predictions are obtained by an end-to-end framework, without explicitly selecting high-level representations. Finally we point out related datasets for experiments and future research directions.

Published
2018
Full Text
View/download PDF

233. Attributes in Multiple Facial Images

Author

Liu, Xudong and Guo, Guodong

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Facial attribute recognition is conventionally computed from a single image. In practice, each subject may have multiple face images. Taking the eye size as an example, it should not change, but it may have different estimation in multiple images, which would make a negative impact on face recognition. Thus, how to compute these attributes corresponding to each subject rather than each single image is a profound work. To address this question, we deploy deep training for facial attributes prediction, and we explore the inconsistency issue among the attributes computed from each single image. Then, we develop two approaches to address the inconsistency issue. Experimental results show that the proposed methods can handle facial attribute estimation on either multiple still images or video frames, and can correct the incorrectly annotated labels. The experiments are conducted on two large public databases with annotations of facial attributes., Comment: Accepted by 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018 Spotlight)

Published
2018

235. Effect of C4+ Irradiation on Mechanical Property of Polycrystalline HIP-SiC at Room Temperature

Author

YI Baiquan;LIU Xudong;HAO Zulong

Subjects
accident-tolerant fuel cladding, hip-sic, ion irradiation, performance characterization, Nuclear engineering. Atomic power, TK9001-9401, Nuclear and particle physics. Atomic energy. Radioactivity, QC770-798
Abstract

Different from other preparation processes, hot isostatic pressing (HIP) has the advantages of improving material density, inhibiting grain growth and avoiding grain orientation, and is often used to prepare polycrystalline 6HSiC materials. Silicon carbide has the characteristics of high melting point, high hardness, corrosion resistance, good thermal conductivity, and will not react with water to produce hydrogen, so it is one of the candidate materials for accidenttolerant fuel (ATF) cladding. In order to investigate the irradiation damage characteristics of polycrystalline 6HSiC prepared by HIP and evaluate the feasibility of silicon carbide material as ATF cladding under HIP process, the experiment took polycrystalline 6HSiC as the research object, and analyzed the properties and structure changes of samples before and after irradiation. In order to prevent the influence of other elements on the experiment, the irradiation ion was C4+, the irradiation dose was 18 dpa and 5 dpa, and a group of non-irradiated samples were set for comparison. The surface characteristics and properties of polycrystalline 6H-SiC before and after ion irradiation were analyzed by SEM, nano indentation, XRD and Raman spectroscopy. The results show that, with the increase of C4+ ion irradiation dose, the relative strength of the Raman curve of HIPSiC sample decreases gradually. Although C4+ irradiation damages the HIPSiC lattice structure to a certain extent, and the sample diffraction peak intensity decreases and shifts, the peak value changes slightly with the increase of the irradiation dose, indicating that the irradiation tends to saturate quickly. After C4+ ion irradiation, the sample material has defects and the nanohardness shows a decreasing trend, but the decreasing degree is not obvious. Therefore, the analysis of ion irradiation, the total number of polycrystalline 6HSiC antiirradiation capacity of the overall hip preparation, has the possibility of being the matrix of the fault in the future.

Published
2022

249. Cortico-amygdalar connectivity and externalizing/internalizing behavior in children with neurodevelopmental disorders

Author

Nakua, Hajer, Hawco, Colin, Forde, Natalie J., Jacobs, Grace R., Joseph, Michael, Voineskos, Aristotle N., Wheeler, Anne L., Lai, Meng-Chuan, Szatmari, Peter, Kelley, Elizabeth, Liu, Xudong, Georgiades, Stelios, Nicolson, Rob, Schachar, Russell, Crosbie, Jennifer, Anagnostou, Evdokia, Lerch, Jason P., Arnold, Paul D., and Ameis, Stephanie H.

Published
2022
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results