National Taiwan University Repository
Not a member yet
222849 research outputs found
Sort by
Expanding the e-commerce market for Japanese groceries in Singapore
No full textThis work defines the e-commerce expansionary strategy for a Singapore-based Japanese grocery provider and critically evaluates the internal constraint and professional guidance to determine how it might be achieved. It considers methodologies to analyze the underserved segmentation needs and demands, review the business landscape and identify the potential market strategy and operation plans. Primary research through questionnaire interviews is conducted as part of this work identified circumstantial factors that correlate to the fluctuating sales. The Massachusetts Institute of Technology (MIT) The Delta Model provided a guiding framework to generating new consumers through a combination of Best Product strategy and Consumers Solution strategy that introduce the otherwise unreachable segment of the targeted population. Besides developing new business through executing the consumer’s solution strategy, it serves as a continual attempt to stay close and relevant to consumers changing needs and demands. To bridge further gaps in the academic framework, the study also considered the scope and activities of marketing and its cost-benefit analysis to deliver a feasible recommendation
Magnetic Resonance Imaging of Rat’s Brain and Spine Using Curved Four-channel Phased Array Coil
No full text磁振造影技術目前已廣泛應用於中樞神經系統的研究中，如解剖影像及擴散張量影像技術等。然而，由於大鼠的頸椎過深，目前商用的磁振造影線圈仍無法提供足夠的影像訊雜比，此減少了磁共振影像於腦脊髓區域研究之可行性。因此，本研究旨在設計與製作一以大鼠中樞神經系統為應用導向之可撓式陣列線圈平台，藉由提昇影像品質來促進磁振造影於腦脊髓之深入研究與探討。 首先，本研究針對大鼠腦脊髓形狀設計並製作了一四通道可撓式陣列線圈。其中，本研究利用幾何重疊與自製的低輸入阻抗前置放大器來降低線圈間之耦合影響。接著，仿體與大鼠之磁共振解剖影像用以評估及驗證此自製線圈之品質。最後，我們將此線圈應用於大鼠腦脊髓之擴散張量影像與神經纖維追蹤影像。 相較於傳統之平面式線圈，可撓式陣列線圈分別可以提供1.3倍與1.45倍的訊雜比增益於仿體與大鼠頸椎之解剖影像實驗中，也同時大幅提昇了脊髓灰白質間的對比雜訊比。此外，更結合寬頻磁振造影技術，以獲得加速11倍的大鼠三維解剖影像及高解析度擴散權重影像。在擴散張量影像的再現性實驗中，利用可撓式陣列線圈所取得的實驗再現性角度差減少為平面式線圈之77%，並且使用可撓式陣列線圈所得到之腦脊髓神經纖維追蹤影像也比用平面式線圈所得之更加完整。 本研究成功地製作了一四通道可撓式陣列線圈平台，除了利用仿體與活體實驗來評估其品質外，本研究更驗證了此線圈於大鼠中樞神經系統之擴散磁振造影的可行性。藉由訊雜比與對比雜訊比之提昇，此可撓式陣列線圈平台有助於更多磁振造影技術於中樞神經系統之研究與應用。Magnetic resonance imaging (MRI) techniques such as anatomical and diffusion tensor imaging (DTI) have been widely applied to investigate the central nervous system (CNS). However, commercial MRI coils could not provide sufficient signal-to-noise ratio (SNR) at the neck due to cervical lordosis (inward curvature), therefore hampering the applications from brain to the spine. The aim of this work was to design and implement a 4-channel curved array coil that provides uniform sensitivity along the CNS, improving studies of the head and spinal column. The 4-channel curved array coil was fabricated on a flexible printed circuit board that could be bent to fit the rat’s contour along the neck while a homemade low input impedance preamplifier eliminates the crosstalk between overlapping coils elements. We evaluated the performance of the coil through anatomical imaging of phantom and rat cervical spine, and finally perform rat cerebrospinal DTI and tractography to show the benefits of a curved array coil. Compared to conventional array coils, the curved array coil offered 1.3- and 1.45-fold SNR gain in phantom and anatomical images of the rat cervical spine respectively. In addition, the contrast-to-noise ratio (CNR) between gray and white matter in spine was alleviated. By combining Wideband MRI technique, the 3D anatomical and high-resolution diffusion weighted images were obtained with a 11-fold acceleration. In reproducibility of DTI, the experimental reproducibility deviation angle acquired by curved array coil was 77% of that by plane array coil. Moreover, the DTI tractography of rat nervous system using the curved array coil was more complete. The 4-channel curved array platform was successfully implemented for rat cerebrospinal MRI. We evaluated its performance by phantom as well as in vivo anatomical imaging and further demonstrated the feasibility of rat cerebrospinal DTI. With improved SNR and CNR, the curved array coil platform could improve or even create new possibilities for biomedical applications in cerebral nervous system
AdaBoost-Based Cecum Recognition System in Accordance with Boston Bowel Preparation Scale
No full text在此篇論文中，我們提出一個可以從各式各樣大腸鏡照片中，自動辨識任一照片是否拍攝到盲腸的系統，藉此減輕大腸鏡醫師檢視大量照片的負擔。 大腸直腸癌在國人近年來癌症的罹患人數及醫療支出費用都名列前茅，值得慶幸的是第0, 1 期的大腸癌經過治療後，五年存活率可達九成以上，但是早期大腸癌並無明顯症狀，必須透過定期做大腸鏡檢查來提早發現，便能有效地提高治療效果，並降低大腸癌致死的風險。 大腸鏡檢查的品質攸關著是否能確實早期發現大腸癌，因此除了病人定期到醫院篩檢，也需要確保大腸鏡檢查擁有一定的品質，經研究發現醫生若在每次檢查都能確實深入到盲腸(盲腸到達率高)，病人罹患大腸癌的機率相對較低，也就是說盲腸到達率是用來評估大腸鏡品質的重要指標。 目前醫院是以人工的方式，交換大腸鏡照片給不同醫師來評估大腸鏡品質，因此我們提出一套自動辨識盲腸的系統，幫助醫師檢視大量照片並計算盲腸到達率，此系統會先利用腸道與糞便的色彩差異評估大腸鏡照片清腸不潔的程度是否影響大腸鏡品質，再針對清腸較乾淨的照片判斷是否含有回盲瓣(Ileocecal Valve,ICV)、三叉紋路(Triradiate Fold)或闌尾口(Appendiceal Orifice)等盲腸的特徵，我們利用各種影像處理的演算法擷取這些盲腸特徵後，再利用機器學習的方式來辨識照片是否有拍攝到盲腸，最後我們在平均辨識準確率上達到94.0%，最高辨識準確率達到96.9%，未來便可利用這套系統判斷醫師是否有確實在大腸鏡檢查中進入盲腸，做為一個公正評估大腸鏡品質的第三方，同時減少人工檢視照片的負擔。In this thesis, we proposed a system which can automatically recognize the cecum image from colonoscopy photos based on the variability of human intestinal. This system can assist doctors to check the colonoscopy photos and reduce the load on doctors. In recent years, the colorectal cancer is the top one cancer on incidence rate and medical expenses in Taiwan. Fortunately, early treatment of colorectal cancer in Tis and T1 can increase the survival rate of patient effectively. However, there is no symptom in the early stage of colorectal cancer. In order to detect the early stage of colorectal cancer, the colonoscopy examination regularly is very important. The colonoscopy quality is closely related to the detection of early cancer. There are some quality indicators for colonoscopy: Cecal Intubation Rate (CIR), Bowel Preparation (BP), Adenoma Detection Rate (ADR), and Withdrawal Time (WT). In this thesis, we focus on CIR and BP. In order to evaluate CIR, doctors need to view great amount of colonoscopy photos. Therefore we propose a cecum recognition system to help doctors to evaluate CIR automatically. The system will assess BP if so bad that we cannot get information and features in the image. Then, the system extracts features of cecum from the images with good BP by image processing, and we use machine learning algorithm to recognize cecum images. Our method achieves the average accuracy rate of 94.0% and the best accuracy rate of 96.9%
The Effects of Agricultural E-commerce Web Quality on Consumers’ Perceived Risk and Purchase Intention
No full text基於Mehrabian和Russell的接受模型，本項研究重點在於探討內部信息源（即網站品質）作為刺激因素，如何影響消費者的反應系統。本研究的目的是測試一個更全面的模型，相關因素包括網站的質量，感知易用性，感知有用性和感知風險，以及購買意願。此外，本文探討網絡質量（其四個構面的網頁設計，私密性/安全性，客戶服務和訂單履行）如何影響感知風險和購買意願。此外，它討論了感知易用性（PEOU）之間的關係，感知有用性（PU）和感知風險。調查結果中提供優質的網絡，網上購物的感知風險，以及在線採購活動的消費者水平的假設關係的證據，也對在線商務和消費者的福利進行了討論。 調查數據以及調查問卷基於結構方程模型（SEM），使用SPSS13.0被用於數據分析。本研究允許檢查對感知風險和行為意圖零售網站的各種部件的不同的影響。這項研究的附加值在於採用相關文獻彌補前人研究的空白，同時也將提供網上零售商設計和維護網站的實用建議，提昇消費者的購買意願。Based on Mehrabian and Russell’s Technology Acceptance Model, this study focuses on internal source of information (i.e. website quality) as stimuli affecting consumers’ response systems. The purpose of this paper is to test a more comprehensive model consisting of website quality, perceived ease of use, perceived usefulness and perceived risk, as well as purchase intention. Further, this paper explores how web quality (4 facets in web design, privacy/security, customer service and fulfillment) influences perceived risk and how these perceptions relate to online purchase intention. Also, it discusses the relationship between perceived ease of use(PEOU),perceived usefulness(PU) and perceived risk. Findings provide evidence of hypothesized relationships among consumers’ levels of web quality, the perceived risks of online shopping, and online purchasing activity. Implications for online commerce and consumer welfare are discussed. Survey data as well as questionnaires will be obtained and structural equation modeling (SEM) will be employed for data analysis by using SPSS13.0. This study allows for examination of the different effects of various components of retail websites on perceived risk and behavioral intentions. This research will add value to the related literature by filling the void of previous research and also will provide practical implications for online retailers on designing and maintaining positive consumer response
Factor analysis of nanowire biosensor and its challenges for future application in market
No full text隨著科技持續的進步與醫療概念的演進，重點式照護診斷的儀器成為研究發展的一個重點，在許多發展中的重點式照護診斷技術當中，奈米線場效電晶體是一個以高靈敏度與快速檢出著稱相當有潛力的一支技術，而心肌冠狀動脈綜合症在現代已成為一個嚴重影響人類壽命與生活品質的一個疾病，因此仔細的探索將奈米線場效電晶體應用於急性冠狀動脈診斷的市場是有必要的，同時這個分析也顯示一些當前以奈米線場效電晶體作為重點式照護檢測技術所需克服的議題。 基於市場分析的結果，本論文採用標準商業CMOS製程製作多晶矽奈米線場效電晶體，用於心血管疾病檢測，為了改善由市場分析所總結出的重要議題，本論文以電腦模擬不同摻雜濃度、通道厚度、氧化層厚度的結果，並以實驗測試所製作之奈米線場效電晶體。As technologies continuously developed, point-of-care testing (POCT) has become an important research direction. Among different developed POCT technologies, nanowire field effect transistor (NW-FET) is one of promising technologies featuring high sensitivity and rapid diagnosis. On the other hand, acute coronary syndrome (ACS) has become a serious cardiovascular disease following developments of human societies. As a consequence, it is intriguing to have a detail market analysis for using NW-FET POCT technology in ACS diagnosis. At the same time, this analysis also shows important issues for the development of NW-FET based POCT technologies. Based on previous research results, commercial-available CMOS processes can be used to implement poly-silicon NW-FET for cardiovascular biomarker diagnosis. Since the market analysis shows several important issues for the development of NW-FET based PCT technologies, in this work, these issues, such as different doping concentration, oxide thickness, and channel depth, are addressed and verified in simulations. In addition, experiments are also employed to examine the fabricated devices. Based on this work, some of the major obstacles of CMOS-based NW-FET biosensors for POCT applications can be raised and evaluated
Features of Mismatch Negativity for Classification of Schizophrenia Patients
No full text聽覺變異刺激實驗可以在刺激後約100毫秒激發失匹配負波（MMN）特徵波形，同時也在標準或異常聲音刺激後約300毫秒呈現特徵波形。在本研究中，我們使用分類器工具來尋找區分精神分裂症患者與正常人的特徵組合。113位精神分裂症患者與95位正常受試者參與聽覺變異刺激實驗並測量事件相關電位（ERP）。我們先計算在標準或異常聲音刺激後約100毫秒及約300毫秒的平均振幅與波峰時間延遲，也使用離散小波轉換來代表MMN ERP的特徵，結果發現在精神分裂症族群與正常族群中，theta頻段ERP波形的標準差與能量存在顯著差異。我們採用支持向量機來比較精神分裂症患者識別的準確性。分類結果顯示，使用訊雜比來表示MMN波形會得到較好的準確分類率，使用線性核函數會得到比使用高斯核函數好的準確分類率。雖然本研究提出的小波時間序列特徵組合並沒有辦法增加分類率，我們認為這是我們第一次嘗試用聽覺MMN ERP來進行精神分裂症患者的區分。An auditory oddball paradigm can elicit the mismatch negativity (MMN) waveform with characteristic waveform at about 100 ms after the stimulus onset. There was also a characteristic waveform at about 300 ms after either standard or deviant sounds presentation. In this study, we used classification tools to find compositions of features in the hope of devising and imaging biomarkers to differentiate between schizophrenia patients and healthy subjects. Specifically event-related potentials (ERP) were measured from 113 schizophrenia patients and 95 healthy controls using an auditory oddball paradigm. The mean amplitudes and peak latencies of the ERP elicited by standard and deviant sounds at about 100 ms and 300 ms were first calculated. We also used discrete wavelet transform to characterize features the MMN ERP’s. We found that the standard deviation and the energy of the ERP waveform in theta band were significantly different between schizophrenia and control groups. Using support vector machine (SVM) as the classifier, we compared the accuracy of identifying schizophrenics. More accurate classification when MMN waveform is described in the signal-to-noise ratio (SNR) unit. Using a linear kernel in SVM gave higher classification accuracy than using a Gaussian kernel. The proposed wavelet time series feature compositions did not improve the classification accuracy. We consider this study is our first attempt to differentiate between healthy and schizophrenic patients using auditory MMN ERP
A Service-oriented and Cross Platform Health Promotion Information System
No full text在台灣,由於過重的人口比例日漸增長,並且有越來越年輕化的趨勢,肥胖儼 然已逐漸成為一個嚴重的公共健康問題。為協助正在進行飲食控制、體重控制亦或 是一般希望維持健康狀態的使用者達到健康促進,此研究中,我們發展了一套資訊 化的健康促進系統--臺大智慧健康照護系統。系統的架構設計主要是利用模組-界 面-控制項的模式進行設計。這套系統目前已經成功的運用於一間公司中。 在新的系統架構設計中,整合並集中了系統內的模組物件,使存取資料庫、使 用者識別及驗證、臨床數值分析引擎的功能集中在單一的元件之中。在這樣的架構 下,可以確保系統的擴充性、系統內的資料一致性以及系統地跨平台能力。本研究 成功的將此架構導入臺大智慧健康照護系統的系統建置之中,並且運用在數種平 台上,包含智慧型手機、個人平板電腦及網站。在使用者訪問系統時,系統使用了 客製化的憑證進行使用者及訪問的辨識,擁有處理網路不穩定的環境的特性。此外, 系統中飲食資料庫的機制則是根據 Web 2.0 的概念進行設計,資料庫的內容也能 由使用者產出,有用的資料量能隨著使用量增加,進而回饋使用者,使得記錄更為 方便。系統上線至 2014 年 5 月 30 日已運行三個月。在這段期間��,相較於過去 該公司所使用的健康照護系統,使用者持續使用的比例提升 25%,而過重的使用 者在這三個月中,平均每月能減少 2Kg。飲食資料庫上線後的一個月則新增了 255 種食物。這樣的結果也顯示使用者願意使用此系統的意願上升,同時增加使用與關 注,有助於體重的控制。同時此架構也顯現了能快速複製出相同的服務的能力,並 擁有未來發展成雲端系統及後續擴充的潛能。Obesity has become an important health issue in Taiwan due to its increasing prevalence among population and the ages of patients are getting younger. In this research, we build a health promotion information system (HPIS), for people who were on diet and to control their weight, or others to keep in shape. The HPIS is designed based on model– view–controller (MVC) pattern with a proposed architecture framework and has been successfully implemented in a company. The architecture centralized the functions of accessing databases, user identification, and the validation of inherited clinical knowledge in a single component. Under this architecture, the system ensures its scalability, extensibility, data consistency, and the capability of cross-platform. This architecture has also been implemented in different devices and platforms, including iPhone, iPad and website. The architecture also supports unstable network environment, which uses Token as session identifier and user validation. Furthermore, a food database mechanism are also proposed in this research. Until May 30, 2014, the system has been launched for three months. Compared with the system that was previously used in a study, the HPIS has retained 25% more users, overweighting people lose approximately 2Kg a month in average. And the food database increases 255 kinds of foods in one month. These results show that the proposed system has the potential of increasing the willingness of users to use it, and through using the system, the health condition of users may receive better control. The proposed framework is also fully capable in extending the service to other companies in the future
Study Of Electrocardiography Power Spectrum
No full text根據行政院衛生署統計，民國九十六年國人十大死因排行榜，心血管疾病死亡率位居第二順位。由於心臟病所引發的心臟血管疾病是無法預期，所以發展一套緊急的警告系統去即時偵測心臟疾病是很重要的，而利用心電圖(Electrocardiogram)來檢測心臟疾病為最簡便有效，且心電圖有價格便宜、即時偵測和容易取得…等優點，因此在臨床上被廣泛使用。 心臟藉由收縮與舒張的運動來傳送血液、氧氣、養份至身體不同的氣官，對人體的循環系統有重大的影嚮力。但隨著心理狀況的不同，心臟跳動的強度和頻率也會有所調整。若仔細觀察，即使人體處於休息狀態也會發現心跳與心跳時間間隔都有微小的差異，即稱為心律變異度。臨床上將心率變異度用來評估自主神經系統的活絡性，其中自主神經系統可以分為交感及副交感神經，影響人體內許多生理變化，故將其發展成為一種生理的指標，可以協助醫生在診斷時作一個參考依據。 傳統利用心率變異度來分析心電圖訊號時，必須先偵測出QRS複合波的位置，再計算兩兩R波峰值之間的時間間隔R-R interval，最後由時域的角度對R-R interval作更進一步的分析。而本文則從不同以往對R-R interval分析的觀點切入，選定既有部份訊號的邊界值，來預測函數後，將其連接起來，使其成為類週期函數，並且連接成質數週期重新取樣後，濾出旁瓣現象，來量測心電圖的真實頻率，我們可以藉由頻譜的倍頻及半高寬分析，來辨別出正常心律及心律不整的人頻譜變化。以及正常心律若出現問題時，會先從高頻的部份開始出現倍頻出現偏移及半高寬會延伸現象，頻譜未來將可以提供早期預測疾病的功能。將來在臨床疾病的治療方面，得知病情的輕重程度。未來將可以此結果建立成資料庫，作為辨別疾病的依據以作為早期治療的預測。According to statistic of Department of Health, Executive Yuan, R.O.C. in 2007, cardiovascular disease has been listed as the second rank of the top ten leading causes of death. Since cardiovascular disease induced heart attack may often occur in unexpected situation, development of an emergent alarm system for early detection of heart dysfunction is important. One most efficient and easy way for detection of early heart dysfunction is the use of electrocardiogram (ECG). ECG has advantages of cheap, real-time detection, and easy to implement which has been widely used in clinics. Via the delivery of blood, heart transfers oxygen and nutrients to various organs and is thus highly influential for circulatory system. To adapt to the variation of physiological conditions, the intensity and frequency of heart beats change with time. Careful observation finds that the time intervals between heartbeats are often different even if the body is at rest. Such as heart rate variability (HRV) has been used to estimate the activity of the autonomic nervous system which can be divided into sympathetic and parasympathetic subsystems both of which can significantly affect the physiological indicator to assist doctors in making diagnostic decisions. cator to assist doctors in making diagnostic decisions. Many studies have used HRV to analyze the ECG signal via studying the QRS complex waveform to determine the time intervals between R-peaks and analyze the R-R intervals from time and frequency domains. Different from the conventional R-R interval based approach, this work introduces new feature variables by computing the spectrogram of the ECG signal waveform. We selected part of the original ECG waveform to get boundary condition and generate predicted function. The new signal is qusai-periodical signal. To clearly analyze the power spectrum, we used the new signal base as qusai-periodical signal to continue to increase the frequent resolution. So that we can clearly distinguish between normal sinus rhythm (NSR) and arrhythmia. The method is validated through experiments on the MIT–BIH databases. The result shows that we can apply this method to distinguish between NSR and arrhythmia. After we filter out sidelobe, now we can get actual frequency without disturbance
Measurement and Analysis of Structural, Electronics Properties of Wurtzite Indium Nitride using X-ray Diffraction, Extended X-ray Absorption Fine Structure and Photoluminescence
No full text本論文的研究主題是量測與分析氮化銦(InN)纖鋅礦晶格結構的晶格常數a，c，鍵結常數 (bond length) ，c/a，以及internal parameter u (b/c)。我們先應用X光繞射求出晶格常數a，c。之後我們再用X光吸收譜來求出纖鋅礦晶格的原子距離(bond length)。我們選擇In K-edge量測分析EXAFS(extended X-ray absorption fine structure)，我們利用X光繞射求出來的晶格常數a，c當作EXAFS擬合的基本參數。 總合HR-XRD,EXAFS的結果,我們成功求出InN半導體的晶格參數a，c，銦-氮的建長，c/a，以及u。我們算出來的u擁有誤差小於10-3Å。 我們再利用photoluminescence量測氮化銦的發光波段.我們發現氮化銦的發光波段在於遠紅外(~ 0.7 eV).造成發光的機制是由導帶的電子躍遷到價帶上面的acceptor state。This master thesis is about studying crystal structure properties of wurtzite structure Indium Nitride using high resolution X-ray diffraction spectroscopy (HR-XRD), and In k-edge Extended X-ray Absorption Fine Structure (EXAFS). The result of HR-XRD does not suffice in resolution since lattice constant results from two different planes differ around 0.13%. We then apply mathematical model to correct the value of the measured lattice constant. The mathematical corrected result of our lattice constant a is ~ 3.53154 to 3.53204 Å, lattice constant c is ~ 5.70437 to 5.70565 Å. Our corrected HR-XRD measurement result has error bar under 10-8 Å. Lattice constant a and c of our samples show significant trend in which whenever the crystal is expanded in basal plane (elongate lattice constant a), the crystal structure will contract in its height (shown in shortened of lattice constant c) and vice versa. We then establish crystal model of wurtzite InN to fit the In k-edge of EXAFS data using the corrected HR-XRD lattice constants. Using combination data from both HR-XRD and EXAFS we could complete the structural measurement of wurtzite InN. We succeed in measuring 1st nearest neighbor (bonds length between In and N), 2nd nearest neighbor (In – In), with these parameters we could then derive wurtzite crystal structure c/a is 1.6152, and internal parameter u (b/c) is 0.3765. We successfully demonstrate the capability of EXAFS and HR-XRD to measure the internal parameter u with error around 10-3A. We notice that despite the lattice different around sample to sample, internal parameter u result shows stability indicating there exists a bonding force to keep the unit tetragonal cell intact by altering their bonding angle. Our study is continued with photoluminescence measurement of InN, with most sample has single peak at 0.68-0.7 eV that correspond to conduction band to acceptor transition