The Chinese University of Hong Kong

About: The Chinese University of Hong Kong is a education organization based out in Hong Kong, China. It is known for research contribution in the topics: Population & Computer science. The organization has 43411 authors who have published 93672 publications receiving 3066651 citations.

Patterns of treatment response in newly diagnosed epilepsy

Abstract: Objective: To delineate the temporal patterns of outcome and to determine the probability of seizure freedom with successive antiepileptic drug regimens in newly diagnosed epilepsy. Methods: Patients in whom epilepsy was diagnosed and the first antiepileptic drug prescribed between July 1, 1982, and April 1, 2006, were followed up until March 31, 2008. Outcomes were categorized into 4 patterns: A) early and sustained seizure freedom; B) delayed but sustained seizure freedom; C) fluctuation between periods of seizure freedom and relapse; and D) seizure freedom never attained. Probability of seizure freedom with successive drug regimens was compared. Seizure freedom was defined as no seizures for ≥1 year. Results: A total of 1,098 patients were included (median age 32 years, range 9–93). At the last clinic visit, 749 (68%) patients were seizure-free, 678 (62%) on monotherapy. Outcome pattern A was observed in 408 (37%), pattern B in 246 (22%), pattern C in 172 (16%), and pattern D in 272 (25%) patients. There was a higher probability of seizure freedom in patients receiving 1 compared to 2 drug regimens, and 2 compared to 3 regimens ( p Conclusions: Most patients with newly diagnosed epilepsy had a constant course which could usually be predicted early. The chance of seizure freedom declined with successive drug regimens, most markedly from the first to the third and among patients with localization-related epilepsies.

Anatomy: simple and effective privacy preservation

Abstract: This paper presents a novel technique, anatomy, for publishing sensitive data. Anatomy releases all the quasi-identifier and sensitive values directly in two separate tables. Combined with a grouping mechanism, this approach protects privacy, and captures a large amount of correlation in the microdata. We develop a linear-time algorithm for computing anatomized tables that obey the l-diversity privacy requirement, and minimize the error of reconstructing the microdata. Extensive experiments confirm that our technique allows significantly more effective data analysis than the conventional publication method based on generalization. Specifically, anatomy permits aggregate reasoning with average error below 10%, which is lower than the error obtained from a generalized table by orders of magnitude.

MaskGAN: Towards Diverse and Interactive Facial Image Manipulation

Abstract: Facial image manipulation has achieved great progress in recent years. However, previous methods either operate on a predefined set of face attributes or leave users little freedom to interactively manipulate images. To overcome these drawbacks, we propose a novel framework termed MaskGAN, enabling diverse and interactive face manipulation. Our key insight is that semantic masks serve as a suitable intermediate representation for flexible face manipulation with fidelity preservation. MaskGAN has two main components: 1) Dense Mapping Network (DMN) and 2) Editing Behavior Simulated Training (EBST). Specifically, DMN learns style mapping between a free-form user modified mask and a target image, enabling diverse generation results. EBST models the user editing behavior on the source mask, making the overall framework more robust to various manipulated inputs. Specifically, it introduces dual-editing consistency as the auxiliary supervision signal. To facilitate extensive studies, we construct a large-scale high-resolution face dataset with fine-grained mask annotations named CelebAMask-HQ. MaskGAN is comprehensively evaluated on two challenging tasks: attribute transfer and style copy, demonstrating superior performance over other state-of-the-art methods. The code, models, and dataset are available at https://github.com/switchablenorms/CelebAMask-HQ.

Quantitative Analysis of Cell-free Epstein-Barr Virus DNA in Plasma of Patients with Nasopharyngeal Carcinoma

Abstract: Using real-time quantitative PCR, cell-free EBV DNA was detectable in the plasma of 96% (55 of 57) of nasopharyngeal carcinoma (NPC) patients (median concentration, 21058 copies/ml) and 7% (3 of 43) of controls (median concentration, 0 copies/ml) Advanced-stage NPC patients had higher plasma EBV DNA levels than those with early-stage disease At 1 month after completion of radiotherapy, plasma EBV DNA was undetectable in 7 of 15 subjects (47%) but remained high in the remaining 8 subjects (53%) Clinical examination revealed that all of the former seven subjects had complete tumor regression, whereas six of the eight latter subjects exhibited evidence of disease persistence or had developed distant metastases These results suggest that quantitative analysis of plasma EBV DNA may be a useful clinical and research tool in the screening and monitoring of NPC patients

Variants in KCNQ1 are associated with susceptibility to type 2 diabetes mellitus.

Abstract: We carried out a multistage genome-wide association study of type 2 diabetes mellitus in Japanese individuals, with a total of 1,612 cases and 1,424 controls and 100,000 SNPs. The most significant association was obtained with SNPs in KCNQ1, and dense mapping within the gene revealed that rs2237892 in intron 15 showed the lowest P value (6.7 x 10(-13), odds ratio (OR) = 1.49). The association of KCNQ1 with type 2 diabetes was replicated in populations of Korean, Chinese and European ancestry as well as in two independent Japanese populations, and meta-analysis with a total of 19,930 individuals (9,569 cases and 10,361 controls) yielded a P value of 1.7 x 10(-42) (OR = 1.40; 95% CI = 1.34-1.47) for rs2237892. Among control subjects, the risk allele of this polymorphism was associated with impairment of insulin secretion according to the homeostasis model assessment of beta-cell function or the corrected insulin response. Our data thus implicate KCNQ1 as a diabetes susceptibility gene in groups of different ancestries.