Prediction Methods in Large-Scale Network Analysis for Neuroimaging Data

Zhu, Ziliang ProQuest Dissertations & Theses The University of 2022 해외박사(DDOD)

Brain functional connectivity data are critical for understanding human brain structure and cognitive disease diagnostics. The underlying genetic architecture behind brain functional connectivity is a critical topic in medical studies, which helps unveil the linkages between genetic variants and brain activity and further understand cognitive diseases and brain disorders. The rapid emergence of large scale imaging studies provides researchers with more opportunities to discover the connections between brain system and genes. However, existing methods in imaging genetics are not sufficient in dealing with the high-dimensional data with complex structure, thus limiting the discovery of biological foundation of neuro-development. Therefore, we developed novel statistical approaches for efficient analysis of imaging genetic data.In the first project, we developed a matrix decomposition based method for denoising and recovering the structure of the subject-wise network based on the assumption of factor model. We decompose the subject networks into two parts: a common low-rank basis and subject-specific loadings on the basis. A matrix L0 penalty problem was formulated to accelerate the algorithm. Meanwhile, to avoid iterative computation of high dimensional matrix, we will select a relatively lower dimension basis in the first step, which is a coarse estimator, and then do a fine-tuning in the second step based on the results in step one. In the simulation study, it showed that our approach outperformed other existing approaches in terms of recovering accuracy and computing speed. We also proved that under mild conditions, the algorithm converges fast in an exponential rate. In the second project, we proposed a matrix regression approach for imaging genetic studies. The proposed regression model includes two steps. In the first step, a marginal screening procedure was used to study the univariate associations between genetic variants (SNPs) and imaging phenotype. The theoretical p-value for the marginal screening step was derived using random matrix theories, and important SNPs were selected based on the univariate associations using knock-off. In the second step, a multivariate regression model with all the important SNPs selected as covariates were fitted, and a penalized optimization problem was solved using Nestrov methods. We studied the theoretical properties of the proposed two-stage algorithm thoroughly and simulation studies supported the efficiency and consistency of the proposed method.In the third project, we established a missing data imputation framework to address the issue of missing image modality in real data. The missingness of some imaging modality is common in real imaging data, which may undermine the statistical power in the prediction and inference. However, inaccurate imputation of the missing modality may lead to bias in prediction. Therefore, we thoroughly studied the performance of imputation approaches, including LASSO and ridge models, under different conditions, and concluded the optimal choice of imputation options under the different settings.

Research on the Impact of Female Investors on Overseas Real Estate Investment : 女性投资者对海外房产投资的影响研究

Zhu, Qingling ProQuest Dissertations & Theses Arizona State Univ 2021 해외박사(DDOD)

Visual Commonsense Reasoning: Functionality, Physics, Causality, and Utility

Zhu, Yixin ProQuest Dissertations & Theses University of Cali 2018 해외박사(DDOD)

Reasoning about commonsense from visual input remains an important and challenging problem in the field of computer vision. It is important because the ability to reason about commonsense, plan and act accordingly, represents the most distinct competence that tells human apart from other animals---the ability of analogy. It is challenging partially due to the absence of the observations of all the typical examples in a given category, in which the objects often present enormous intra-class variations, leading to a long-tail distribution in the dimensions of appearance and geometry. This dissertation focuses on four largely orthogonal dimensions---functionality, physics, causality, and utility---in computer vision, robotics, and cognitive science, and it makes six major contributions. We rethink object recognition from the perspective of an agent: how objects are used as "tools" or "containers" in actions to accomplish a "task". Here a task is defined as changing the physical states of a target object by actions, such as, cracking a nut or painting a wall. A tool is a physical object used in the human action to achieve the task, such as a hammer or a brush, and it can be any daily objects which are not restricted to conventional hardware tools. This leads us to a new framework---task-oriented object modeling, learning and recognition, which aims at understanding the underlying functions, physics and causality in using objects as tools in various task categories. We propose to go beyond visible geometric compatibility to infer, through physics-based simulation, the forces/pressures on various body parts as people interact with objects. By observing people's choices in videos, we can learn the comfort intervals of the pressures on body parts as well as human preferences in distributing these pressures among body parts. Thus, our system is able to "feel", in numerical terms, discomfort when the forces/pressures on body parts exceed comfort intervals. We argue that this is an important step in representing human utilities ---the pleasure and satisfaction defined in economics and ethics (e.g., by the philosopher Jeremy Benthem) that drives human activities at all levels. We propose to go beyond modeling the direct and short-term human interaction with individual objects. Through accurately simulating thermodynamics and air fluid dynamics, our method can infer indoor room temperature distribution and air flow dynamics at arbitrary time and locations, thus establishing a form of indirect and long-term affordance. Unlike chairs in a sitting scenario, the objects (heating/cooling sources) that provide affordance do not directly interact with a person. Instead, the air in a room serves as an invisible medium to pass the affordance from an object to a person. We coin this new form of affordance as intangible affordance. By fusing functionality and affordance into indoor scene generation, we propose a systematic learning-based approach to the generation of massive quantities of synthetic 3D scenes and numerous photorealistic 2D images thereof, with associated ground truth information, for the purposes of training, benchmarking, and diagnosing learning-based computer vision and robotics algorithms. We present four case studies on integrating forces and functionality in object manipulations in the field of robotics, showcasing the significance and benefits of explicit modeling of the functionality in task executions. We introduce an intuitive substance engine (ISE) model employing probabilistic simulation, which supports the hypothesis that humans infer future states of perceived physical situations by propagating noisy representations forward in time using approximated rational physics.

Learning affordances for maximum distance throws in the context of learning to throw

Zhu, Qin Indiana University 2008 해외박사(DDOD)

By hefting objects in the hand, people are able to judge the object affordance (the optimally weighted object at a given graspable size) for maximum distance throws (Bingham at al., 1989; Zhu & Bingham, in press). This affordance corresponds to a relation between object size and weight and distances of throws, that is, a single valued function (distance) in two variables (size and weight). The present study first explored whether this affordance could be learned with the acquisition of throwing skill, and second attempted to identify whether the acquisition of this affordance is a type of function learning (Busemeyer & McDaniel, 1997) or the acquisition of a smart perceptual mechanism (Bingham, et al., 1989). 24 unskilled adult throwers were asked to heft 48 objects of different sizes and weights, and to judge their affordances for the maximum distance throws. A month long intensive practice of throwing was then administered, for which participants were divided into 4 groups so that 3 groups practiced throwing with vision using 3 prescribed sets of 6 objects each (constant size, constant weight or constant density) but the 4th group without vision using the set in constant density. After practice, hefting judgments, throwing and then hefting judgments again were tested with the full set of 48 objects. The results showed that participants, were unable to perceive the affordance before practice, however, as throwing skill was acquired through practice and the visual perception of throwing distance was provided, their sensitivity to the affordance improved independently of the prescribed set of objects, and finally, accurate perception of the affordance was acquired. Study also confirmed that only object weight affected the dynamics of throwing to determine the release velocity although the throwing distance was determined jointly by object size and weight. The results indicated that the affordance was perceived using a smart perceptual mechanism that was also acquired as participants learned to throw.

The Effects of Online Review Ratings: A Case Study of the Hotel Industry

Zhu, Zhu ProQuest Dissertations & Theses Boston College 2023 해외박사(DDOD)

Online reviews have gained importance for consumers when shopping for experience goods. This dissertation documents the impact of Tripadvisor.com reviews on the hotel industry. In the first chapter, I investigate the causal impact of Tripadvisor review ratings on hotel performance via a regression discontinuity design. The results indicate that a 1-point increase in review rating leads to a 1.6% increase in revenue, a 1% increase in bookings, and a 0.4% to 0.6% increase in prices. Furthermore, the impact on bookings has increased over time. In the second chapter, I evaluate the welfare impact of Tripadvisor review ratings in providing information about quality. I develop a structural model of hotel demand and supply that takes price endogeneity and capacity constraints into consideration. Counterfactual experiments reveal that the removal of Tripadvisor from the status quo results in per-capita consumer surplus loss ranging from $0 to $5.8, with a more significant decrease in consumer surplus when prior knowledge about quality is less accurate. Hotels with higher quality than expected absent reviews benefit from review ratings, while the opposite is true for others. In the third chapter, I analyze the relative influence of Tripadvisor ratings on chain-affiliated and independent hotels and evaluate the value of Tripadvisor ratings compared to chain brands using the methodology developed in previous chapters. I find there is no significant difference in the effect of rating rounding on occupancy rates for chain-affiliated hotels versus independent hotels. Counterfactual experiment results suggest that despite chain brands providing value to consumers, Tripadvisor ratings provide additional value of about $0 to $4 per capita. In scenarios where Tripadvisor was not present, Chain-affiliated hotels benefit from brand affiliation while independent hotels are harmed.

An Acoustic Study of English Diphthongs Produced by Chinese and Korean Learners of English

Zhu Kunling 청주대학교 대학원 2016 국내박사

It is widely accepted in the field of Second Language Acquisition (SLA) that one’s first language (L1) exerts certain influences on the phoneme acquisition of a second language (L2). The present research aims at investigating the effect of L1 diphthong inventory on the acquisition of the English falling diphthongs by probing deeper into the acoustic features of each diphthong produced by learners with different L1 backgrounds: Mandarin Chinese and Korean which have quite different diphthong inventories. Twenty college students from China and twenty from Korea were recorded to get enough tokens for each diphthong, and six American native speakers were recorded as the comparison group. Every diphthong produced by the three groups was compared based on the acoustic parameters of duration, formant rate of change (F1 ROC and F2 ROC), trajectory length (TL) and spectral ROC. Furthermore, through the analysis of the acoustic features of the diphthongs, the effect of different task types was also explored. It was revealed that the two L2 speaker groups differ from the American group more or less for most of the diphthongs. Those produced by the Chinese speakers tend to be significantly shorter in duration and TL while smaller in formant ROC and spectral ROC. Most diphthongs produced by the Korean group are closer to those of the Americans in many acoustic parameters. The Korean participants, whose native language does not possess any falling diphthong, outperform their Chinese counterparts who have similar diphthongs in their mother tongue. The current research is of theoretical as well as practical significance. It provides some new insights into the nature of diphthong phonological transfer from Chinese and Korean to English. The similar diphthong inventories between Chinese and English cause long-term difficulties for the learners while the absence of falling diphthongs from Korean fosters the learning process. It also has pedagogical implications to English pronunciation teaching in China and Korea. If the teachers get a better understanding of the specific phonological influence of their L1s, they will be better equipped with the ability of helping students grasp the phonological system of English. 제2언어 습득분야에서 일반적으로 용인되는 사실은 화자의 모국어가 제2언어의 음소를 습득하는 과정에 어느 정도 영향을 끼친다는 점이다. 본 연구의 목표는 상이한 이중모음 체계를 가지고 있는 중국어와 한국어의 화자가 미국 영어의 이중모음을 발화하는데 해당 모국어가 어떠한 영향을 미치는 지 살펴보는 데 있다. 중국어 화자 20명과 한국어 화자 20명을 모집해서 발화실험에 참여시켰고 미국 영어 화자 6명이 대조군으로 실험에 참여하였다. 중국어, 한국어, 미국 영어 화자가 발화한 이중모음을 길이, 음형대 변화, 음형대 궤적의 길이, 스펙트럼의 변화와 같은 음향적 특징을 추출해서 비교하였다. 더불어 이중모음의 음향적 특징을 분석해서 이중모음이 발화된 다른 환경의 효과 또한 분석하였다. 음성분석 결과에 따르면 중국어 화자와 한국어 화자가 발화한 영어의 이중모음은 서로 성격이 다른 것으로 나타났다. 미국 영어 화자가 발화한 이중모음은 중국인 화자와 비교했을 때 이중모음의 길이와 음형대 궤적의 길이 측면에서 통계적으로 유의미하게 작게 나타났고 음형대와 스펙트럼의 변화에서도 많은 차이가 나타났다. 하지만 한국어 화자가 발화한 이중모음은 미국 영어 화자가 발화한 이중모음과 비슷한 양상을 보였다. 이러한 양상에 대한 까닭은 한국어에는 미국 영어에 있는 하향이중모음이 없는 반면 중국어에는 영어와 유사한 하향이중모음이 있기 때문이다. 그래서 한국어 화자가 발화한 미국 영어의 이중모음은 상대적으로 중국인이 발화한 이중모음에 비해서 미국영어의 이중모음과 성격이 유사했다고 판단된다. 본 연구는 이론적인 측면과 실용적인 측면에서 중요성을 지니고 있다. 본 연구의 결과는 중국어와 한국어가 영어 간에 나타나는 이중모음의 음소 전이 성격에 대해서 새로운 통찰력을 제공할 수 있다고 생각한다. 중국어와 미국 영어의 유사한 이중모음의 체계는 영어를 습득하는 중국인 화자에게 어려움을 야기시키지만 한국어와 미국영어의 상이한 이중모음의 체계는 영어를 배우는 한국인 화자에게 중국어에서 발견되는 어려움 같은 것을 발견되지 않고 오히려 영어 음소를 학습하는데 있어서 촉진역할을 한다고 여겨진다. 제2외국어를 가르치는 교사들이 이러한 언어 간에 음소전이의 성격에 대해서 충분히 이해할 수 있다면 영어의 발음을 가르치는 데 있어서 학생들의 발음을 향상시킬 수 있는데 도움이 된다고 여겨진다.

가시적 비밀 조각 모자이크 이미지 방식의 개선된 정보 은닉 기법

Zhu Lin Chosun University, Graduate School of Chosun Unive 2012 국내석사

We propose a new method for information hiding based on secret fragment visible mosaic image. This algorithm has the been developed for the application of color transfer to generate the mosaic image hiding the information. Since the RGB color space has strong correlation between three color channels, we must modify all color channels in tandem. Thus it complicates any color modification process. Another problem in this color space is the size of bit stream of recovery information of secret image block sequence. Hence embedding this large bit stream in the mosaic image reduces the quality of the resulting image. In this paper, a new way to solve these problems using the color transfer technique in correlated color space was proposed. The pixel value is considered as a three dimensional stochastic variable and an image a set of samples. Color transfer process is achieved by the geometrical transformations, such as translation, scaling and rotation. The correlation between the color spaces is measured and further used for the color transfer process by studying the covariance between the color space. In order to get more robust encryption and lower bit rate requirement, the combination of logistic map and Chebyshev map is used to generate the encryption sequence of the secret image. 이 논문에서는 가시적 비밀 조각 모자이크 이미지를 기반으로 새로운 정보 은닉 방법을 제안한다. 이 알고리즘은 정보를 숨겨 있는 모자이크 이미지를 생성하는 컬러 전송의 응용 프로그램에 대한 개발되었다. RGB 색 공간은 세 컬러 채널 간의 강한 상관관계를 가지고 있기 때문에, 탠덤의 모든 색상 채널을 수정해야한다. 따라서 어떤 색상 수정 과정이 복잡해지는 문제가 있다. 이 색상 공간에서 또 다른 문제는 비밀 이미지 블록 시퀀스의 복구 정보의 비트 스트립의 크기이다. 따라서 모자이크 이미지에 큰 비트 스트립을 포함하는 것은 결과 이미지의 품질을 악화시킨다. 본 논문에서는 서로 관련적인 색상 공간에서 색상 전송 기술을 사용하여 이러한 문제를 해결하는 새로운 방법을 제안한다. 픽셀 값은 입체 확률 변수와 이미지 샘플의 집합으로 간주된다. 색상 전송 프로세스는 번역, 스케일링 및 회전으로 구성된 기하학적인 변환에 의해 이루어진다. 색상 공간중의 상관관계는 공분산을 연구하여 측정 및 색상 전송 프로세스에 사용된다. 강력한 암호화 및 낮은 비트 레이트 요건을 얻기 위해서 logistic map 하고 chebyshev map의 조합이 사용되어 비밀 이미지 암호화 시퀀스를 생성하게 된다.

한중 FTA하에서 한중 농산물 무역에 관한 연구

ZHU HONGSONG 한국해양대학교 대학원 2022 국내석사

Since the 21st century, the integration of the world economy has become more pronounced, more and more countries are getting involved. In addition, in accordance with the flow of the times, the world economy has moved according to the order of the WTO. However, as trade disputes between member states began to emerge, some countries signed free trade agreements (FTAs) to bypass the difficulties of multilateral agreements at the WTO and facilitate trade liberalization. Currently, many countries have signed FTAs under the WTO system. Accordingly, China and Korea also conducted FTA negotiations with China's proposal in 2002 in line with global changes. The greatest goal of FTA negotiations is to maximize the interests of bilateral trade countries. After Korea and China formally established diplomatic ties in 1992, the political, economic, and cultural trade between the two countries developed rapidly. Agricultural trade also continued to increase, In particular, the interdependence of agriculture has increased. According to statistics from the Ministry of Commerce of China, China has developed into the second largest importer of agricultural products from South Korea after the United States. Therefore, in the process of signing the Korea-China FTA, the negotiation in the field of agricultural product trade has become the most concerned issue for Korea. Agricultural trade became a sensitive issue during the Korea-China FTA negotiations, but after 14 rounds of negotiations from 2012 to 2015, it passed the Korean National Assembly in November 2015. On December 20 of the same year, the Korea-China FTA was implemented in earnest, opening up agricultural markets between Korea and China. It is currently the 7th year of the Korea-China FTA. Under the framework of the Korea-China FTA, this paper studies the impact of the conclusion of the FTA on the agricultural trade between the two countries and the impact of the Korea-China agricultural trade on the Korean economy. In addition, a comparative analysis was carried out on the current situation of agricultural trade between South Korea and China, the changes of agricultural trade in South Korea and China FTA, the scale, type and structure of trade in agricultural products between South Korea and China. Analyze the effect of tariff concessions on agricultural products of both Korea and China. Lastly, we will consider cooperation measures for the development of agricultural trade between Korea and China, and establish future trade strategies. KEY WORDS: Korea-China FTA, agricultural product trade, tariff concession effect 21세기 이래 세계경제의 일체화 현상이 더욱 현저해지고, 점점 더 많은 국가가 참여하게 되었다. 또한, 시대의 흐름에 순응하여 세계경제는 WTO의 질서에 따라 움직이게 되었다. 그러나 회원국 간의 무역분쟁 문제가 불거지기 시작하면서 WTO의 다자간 합의의 어려움을 우회하여 무역자유화를 추진하기 위해 여러 나라가 FTA(Free Trade Agreement)를 체결하였다. 현재 많은 국가들이 WTO 체제하의 FTA를 체결하고 있다. 이에 중국과 한국도 세계적인 변화에 발맞춰 2002년 중국의 제안으로 FTA 협상을 진행 하였다. FTA 협상의 최대 목표인 양자 무역 국가의 이익 극대화이다. 1992년 한중 양국이 정식으로 수교한 후 두 나라 사이의 정치, 경제, 문화 무역이 급속히 발전하고, 농산물무역도 지속적으로 증가하였다. 특히, 농업상의 상호의존도가 증가하였는데, 중국 상무부 통계에 따르면 중국은 이미 한국에 있어 미국에 이어 두 번째로 큰 농산물 수입국으로 발전하였다. 이에 따라 한중 FTA 체결과정에서 농산물무역 분야 협상은 한국의 최대 관심사로 떠올랐다. 한중 FTA 협상 과정에서 농산물무역은 민감한 이슈가 되었으나 이후 2012년부터 2015년까지 14차 협상을 거쳐 2015년 11월 한국 국회를 통과했고, 같은 해 12월 20일 한중 FTA가 본격 시행되어 한중 양국은 농산물 시장을 개방하였다. 현재 한중 FTA 이행 7년째이다. 한중 FTA 하에서 본 논문은 FTA 체결이 양국 농산물무역에 미치는 영향과 한중 농산물무역이 한국 경제에 미치는 영향에 대하여 연구한다. 또한, 한중 양국의 농산물무역 현황, 한중 FTA 농산물무역 변화, 한중 양국의 농산물무역 규모, 무역종류와 구조를 비교 분석하며, 한중 양국 농산물의 관세 양허 효과를 분석한다. 마지막으로 한중 농산물무역이 발전하기 위한 협력 방안을 고찰하며, 향후 무역 전략을 세운다 . 키워드: 한중 FTA , 농산물 무역, 효과

Controlled Release and Bioavailability of Poorly Water Soluble Drugs Using Nanoemulsifying Solid Dispersion

Piao, Zong Zhu 강원대학교 대학원 2009 국내박사

In the previous study, the solid self-nanoemulsifying solid dispersion system was prepared using the hot melting method. The optimal formulation was consisted of ITZ, poloxamer 407, oleic acid, glycerin, Brij 98, Primojel, adsorbent and butylated hydroxyl toluene (BHT). The objective of the present study was to prepare a solid dosage form containing solid dispersion system as well as to evaluate its in vitro dissolution and in vivo study. The core tablets consisting of the SD were formed by direct compression, and then coated with polymer. The in vitro release characteristics of tablet were evaluated in the enzyme-free simulated gastric fluid (SGF, 1.2) and simulated intestinal fluid (SIF, pH 6.8) and further investigated the stability at various storage conditions. In vivo study of tablets (containing 55 mg ITZ) was carried out in eight healthy male human, the eight volunteers were randomly divided into two groups for a balance 2×2 two-way crossover design. Concentration of ITZ in the plasma was determined using HPLC with UV detector. Although the test tablets were low dose, the pharmacokinetic parameters such as AUC_(0-72), C_(max) and T_(max) were almost similar to commercial Sporanox^(??) capsules (containing 100 mg ITZ) and insignificant sequence effect was found for all of the pharmacokinetic parameters. Hence, it was concluded that the solid self-nanoemulsifying SD system could be considered as a valuable alternative for oral dosage forms of ITZ. The primary objective of the present study was: (i) to improve the solubility and dissolution rate of a poorly water-soluble drug, valsartan (VAL) and (ii) to validate a sensitive method of determining VAL concentration in human plasma samples using high performance liquid chromatography (HPLC) combined with ultraviolet (UV) detection. The ternary solid dispersions (SDs) with poloxamer 407 as a carrier and Aeroperl^(??) 300 as an adsorbent were prepared using the hot melt method. The physicochemical properties of solid dispersed VAL were analyzed by using differential scanning calorimetry (DSC), scanning electron microscopy (SEM), powder X-ray diffraction (PXRD) and Fourier transform infrared spectroscopy (FT-IR). The particle size of VAL in dissolution medium was measured by dynamic light scattering (DLS). It was shown that VAL was an amorphous state in the system and the absence of any significant interaction with other additives. The dissolution rate of SDs was markedly enhanced compared to pure VAL in enzyme-free simulated gastric fluid (SGF), attaining drug releases of up to 80% after 2 hr. The SD loaded controlled release tablets were prepared by direct compression and further investigated the stability at various storage conditions and the pharmacokinetic parameters of VAL were compared to commercial Diovan^(??) tablets. After a plasma simple protein precipitation using methanol, the analytes were separated on a Phenomenex Luna C18 column using 42% acetonitrile with 15 mM potassium dihydrogenphosphate in water (pH 2.0; adjusted with phosphoric acid) as the mobile phase at a flow rate of 1.2 ml/min. The standard calibration curve constructed in the concentration range of 50-2000 ng/ml showed good linearity (r² >0.9997). Spironolactone was used as an internal standard (I.S). VAL and IS eluted at 10.25 and 12.17 min, respectively. The intra-day and inter-day precision and accuracy were satisfactory with relative standard deviations of less than 15%. No interference peaks or matrix effects were observed in human plasma. VAL concentration in human plasma was well established following a single 80 mg oral dose to eight healthy volunteers. The current determination of VAL concentration by protein precipitation using methanol followed by analysis using HPLC with UV detection was rapid and sensitive, and provide an alternative to the analysis of VAL by studying its clinical applications. The objective of the present study was to improve the solubility and dissolution rate of a poorly water-soluble drug, itraconazole (ITZ), using a solid self-nanoemulsifying (SNE) solid dispersion system. The solid dispersion (SD) systems were prepared by the hot melt method. The physicochemical properties of solid dispersed ITZ were characterized by using differential scanning calorimetry (DSC), scanning electron microscopy (SEM), powder X-ray diffraction (PXRD) and Fourier transform infrared spectroscopy (FT-IR). It was shown that ITZ was an amorphous state in the SD system and the absence of any significant interaction with other additives. The dissolution rate of ITZ from SD was evaluated using USP dissolution method II (paddle method) in the enzyme-free simulated intestinal fluid (SIF). The solid self-nanoemulsifying SD system was markedly enhanced dissolution rate of ITZ. The increased dissolution rate was attributed to increased solubility and improved wetting of ITZ crystals. The solid SNE solid dispersion system was also useful at improving the dissolution rate by forming a microemulsion droplets or dispersible particles within the gastrointestinal (GI) tract. Droplet size of ITZ emulsion measured by dynamic light scattering (DLS) and transmission electron microscope (TEM) was constant in the simulated intestinal fluid (SIF, pH 6.8) after 60 min of dissolution process. The release rate of solid SNE solid dispersion system was markedly enhanced compared to pure ITZ in the SIF. The method could simply and easily scaled up and have a general applicability to many poorly water-soluble drug entities in their oral routes. 본 논문의 목적은 난용성 약물인 Itraconazole (ITZ)과 Valsartan (VAL)를 함유한 새로운 self-emulsifying solid dispersion system을 구축하여 약물의 dissolution rate와 bioavailability를 제고하고 평가 하는 것이다. Chapter I 에서는 poorly water-soluble weakly basic ITZ를 모델약물로 선정하고 solid self-nanoemulsifying solid dispersion를 hot melt method으로 만든 후 평가하였다. ITZ를 함유한 Solid dispersion는 SIF (pH 6,8)에서 시판품 Sporanox^(??) 캡슐과 비하여 높은 용출율을 나타내었으며 DSC, PXRD and SEM 등 기계장비를 통하여 약물이 amorphous state로 존재하고 있음을 입증하었으며 또한 약물과 additives는 서로 interaction이 없이 안정함을 FT-IR로 증명하였다. Solid dispersion는 gastrointestinal (GI) tract에서 자율적으로 nanoemulsion droplet를 형성하었으며 DLS and TEM등 기계장비로 그 size를 측정한 결과 용출 1시간후 약 260 nm 었다. Chapter II 에서는 self-nanoemisifying solid dispersion을 로딩한 정제를 만들고 용출율과 생체이용율을 시판품과 비교평가 하였다. Solid dispersion를 loading 한 정제 (containing 55 mg ITZ)는 enzyme-free simulated gastric fluid (SGF) and simulated intestinal fluid (SIF)에서 높은 용출율을 나타내었으며 비록 낮은 dose이지만 AUC_(0-72), C_(max) and T_(max) 와 같은 pharmacokinetic parameters는 시판품 Sporanox^(??) capsules (containing 100 mg of ITZ)과 통계학적으로 유사하였다. Chapter III에서는 poorly water-soluble weakly acidic VAL를 모델약물로 선정하고 solid dispersion를 로딩한 controlled release (CR) matrix tablet를 만들고 용출율과 생체이용율을 평가하였다. VAL를 함유한 Solid dispersion는 SGF (pH 1.2)에서 시판품인 Diovan^(??) tablet과 비하여 높은 용출율을 나타내었으며 DSC, PXRD and SEM 등 기계장비를 통하여 약물이 amorphous state로 존재하고 있음을 역시 입증하었다. solid dispersion를 로당한 controlled release (CR) matrix tablet는 직타법으로 만들었으며 약물은 서방적인 방출 양상을 나타내었다. A rapid and reproducible HPLC-UV detection method was created using a one-step solvent extraction process for the determination of VAL concentration in human plasma for in vivo pharmacokinetic studies. 본 연구에서 개발한 새로운 silid nanoemlisifying solid dispersion system은 기타 난용성 약물의 생체이용율의 개선에도 이용될 수 있을 것으로 기대된다.

Empirical Analysis of Factors Affecting Consumers' Continued Intention to Use Online Platforms

ZHU LIN 경희대학교 대학원 2023 국내박사

As the COVID-19 pandemic forced lifestyles to shift from offline to online globally, Internet platforms developed rapidly and gradually became an integral part of daily life. This study explores the factors that influence MZ-era consumers – the generation that grew up with Internet technology – to continue using Internet platforms after the COVID-19 pandemic. We investigate the impact of the perceived quality and risks of Internet platforms on consumers’ trust and privacy concerns, as well as the moderating effect of trust in the Personal Information Protection Act. We conducted online focus group interviews with MZ-era consumers who grew up using Internet technology and prepared and distributed questionnaires after deriving variables by organizing and coding the content of the interviews for analysis. SPSS 26.0 and AMOS 26.0 were used to analyze the data from the 455 valid collected questionnaires. The empirical results showed that the perceived quality characteristics of Internet platforms in terms of their information accuracy, service convenience, and system security had a significant positive effect on trust in such platforms. Further, service convenience and system security were found to have a significant negative effect on privacy concerns. Information collection risk and GPS location tracking risk among the perceived risks of Internet platforms had a significant negative impact on trust in platforms, while information collection risk and payment risk had a significant positive impact on privacy concerns. Platform trust had a significant negative effect on privacy concerns, and platform trust had a significant positive effect on continued intention to use. Privacy concerns had a significant negative effect on continued intention to use. In addition, a significant moderating effect of trust in the Personal Information Protection Act was found between perceived payment risk and privacy concerns on platforms. The results of the study allow us to identify specific factors that influence consumers’ intentions to continue to use Internet platforms: enhancing consumers’ trust in platforms by improving information accuracy, service convenience, and system security; reducing the risk of information collection, GPS location tracking risks, and payment risk; and reducing consumers’ privacy concerns. It also recognizes the importance of consumers’ trust in the Personal Information Protection Act under national policy and legal regulations. Corresponding operational recommendations are provided for the management of Internet platforms.