1 |
Discriminant analysis for cardiology ultrasound in left ventricleChen, Jie-Min 05 July 2011 (has links)
This study investigates use of echocardiography to assess the related issues about whether the heart function of a subject examined is normal or not. Two-dimensional echocardiography can make the heart inspection, and provide very detailed informations for each part of the cardiovascular structures with a high degree of accuracy. Many studies indicated that the systolic and diastolic function with left ventricle of heart disease in patients was poorer than those of normal patients. Therefore it is of interest to study the systolic and diastolic function for examining whether there are heart problems. In this work, the data is the gray-scale values of left ventricular static ultrasound images. The gray-scale differences between systolic and diastolic period, are used to assess whether the patient suffers from the heart diseases or not. Here, we use factor analysis to simplify and select the crucial factors, namely the function in different area of the left ventricle. Finally, linear and quadratic discriminant analyses are used to distinguish the normal and the abnormal subjects.
|
2 |
Comparison of Classification Effects of Principal Component and Sparse Principal Component Analysis for Cardiology Ultrasound in Left VentricleYang, Hsiao-ying 05 July 2012 (has links)
Due to the association of heart diseases and the patterns of the diastoles and systoles of heart in left ventricle, we analyze and classify the data gathered form Kaohsiung Veterans General Hospital by using the cardiology ultrasound images. We make use of the differences between the gray-scale values of diastoles and systoles in left ventricle to evaluate the function of heart. Following Chen (2011) and Kao (2011), we modified the way about the reduction and alignment of the image data. We also add some more subjects into the study.
We treat images in two manners, saving the parts of concern. Since the ultrasound image after transformation to data form is expressed as a high-dimensional matrix, the principal component analysis is adapted to retain the important factors and reduce the dimensions. In this work, we compare the loadings calculated by the usual principal and sparse principal component analysis, then the factor scores are used to carry out the discriminant analysis and discuss the accuracy of classification. By the statistical methods in this work, the accuracy, sensitivity and specificity of the original classifications are over 80% and the cross validations are over 60%.
|
3 |
THE ROLE OF PSYCHOLOGICAL DISCREPANCY IN BELIEF CHANGE: TESTING FOUR MODELS WITH A SINGLE MESSAGEHuang, Luling January 2020 (has links)
In belief change, message discrepancy is the difference between the belief position advocated in a message and a message receiver’s initial belief position. Psychological discrepancy is the message discrepancy experienced by the receiver. Existing literature had assumed that a high level of psychological discrepancy discounted the weight of a message, which could make the message less effective. However, there were three alternative assumptions about the role of psychological discrepancy. The problem that this dissertation examined was: Does psychological discrepancy affect the weight of a message only, affect the scale value of the message only, affect neither, or affect both? To find out which of these four assumptions was more plausible, this dissertation derived competing hypotheses based on four mathematical models either through an analytic proof alone or a combination of an analytic proof and a computational approximation. This dissertation tested these hypotheses in an experiment with a 3 (high vs. moderate vs. low message scale value) ´ 3 (high vs. moderate vs. low upper bound) between-subjects design (N = 448 Mechanical Turk workers). The results showed that the weight-discounting model had the most supported hypotheses and fit the data the best, which indicated that psychological discrepancy affected the weight of a message only. This dissertation improves the understanding of the mechanism that leads to the outcomes posited by discrepancy models in persuasion research and provides additional empirical evidence for the scale value constancy assumption in information integration theory. / Media & Communication
|
4 |
Comparison of Discrimination between Logistic Model with Distance Indicator and Regularized Function for Cardiology Ultrasound in Left VentricleKao, Li-wen 08 July 2011 (has links)
Most of the cardiac structural abnormalities will be examined by echocardiography. With more understanding of heart diseases, it is commonly recognized that heart failures are closely related to left ventricular systolic and diastolic functions. This work discusses the association between gray-scale differences and the risk of heart disease from the changes in left ventricular systole and diastole of ultrasound image. Owing to the large dimension
of data matrix, following Chen (2011), we also simplify the influence factors by factor analysis and calculate factor scores to present the characteristics of subjects.
Two kinds of classification criteria are used in this work, namely logistic model with distance indicator and discriminant function. According to Guo et al. (2001), we calculate the Mahalanobis distance from each subject to the center of normal and abnormal group, then use logistic model to fit the distances for classification later. This is called logistic model with distance indicator. For the discriminant analysis, the regularized method by Friedman (1989) for estimation of covariance matrix is used, which is more flexible and can improve the covariance matrix estimates when the sample size is small. As far as the
cut-point of ROC curve, following the approach as in Hanley et al. (1982), we find the most appropriate cut-point which has good performances for both sensitivity and specificity under the same classification criteria. Then the regularized method and the cut-point of ROC curve are combined to be a new classification criterion. The results under the new
classification criterion are presented to classify normal and abnormal groups.
|
5 |
Value networks in the Colombian small-scale textile artisanal sectorSANCHEZ ALDANA, ELIANA MARIA January 2013 (has links)
In this research, I report and analyze the experiences of people working with small-scale enterprises in the Colombian textile artisanal sector for understanding how artisanal products are delivered to national and international markets. The data was collected by interviewing design managers and managers currently working with several of these small-scale enterprises. This paper was done considering four important features: firstly, the enormous potential that the Colombian textile industry once held back in the 1960’s; secondly, Colombia’s rich and multiple artisanal creations as expression of culture and idiosyncrasy that show the country’s ample geography and culture; thirdly, the characteristics of the Colombian economy and industry which is mainly composed of micro and small enterprises; and lastly , the lack of research on small-scale value networks. In this research the idea of linear value chains is superseded by the notion of value networks, in which relationships, activities and actors are interconnected to co-create value. In this setting, culture, people and sustainability are the key factors to reach differentiation. By learning from these experiences and analyzing practices in small-scale enterprises I can suggest that, new approaches that benefit and reinforce the characteristics of the small and medium scale enterprises must be built as an opportunity for improving the lives of people in developing countries. Additionally, relationships and the order of the production and marketing flows within the value networks depend on the manager approach. / Program: Textilt management med inriktning styrning av textila värdekedjor
|
6 |
The judgment study for attributes and problems of place marketing of Kaohsiung in pluralistic societies¡ÐThe applications and comparisons of multiple-attribute weight measurement methods.Dai, Ya-ming 06 July 2007 (has links)
In the current pluralistic societies, public policy-makings had to take into account various stakeholders as well as the social values. The research, based on the integrated public affairs management¡]PAM¡^ framework (Wang et al., 2004a) and place marketing strategy (Kotler, 1999), defined the applicable operational variables which includes three major factors of economical, social and political, and government, and nine derivative attributes, two situations of direct three-links between Taiwan and Mainland China, and status quo are used as the scenario. Altogether they will influence the stakeholders as decision-makers while deciding if they should invest, live, or work in Kaohsiung.
For research design, we employed three major multiple-attribute theories or methods for individual stakeholders¡¦information processing and integration, Information Integration Theory¡]IIT¡^, Analytical Hierarchy Process¡]AHP¡^, Simple Multi-Attribute Rating Technique¡]SMART¡^, and Rating. In particular, the way the information is integrated cognitively is empirically examined instead of assumed with the averaging model of the IIT (Anderson, 1991a), and the scale values (equivalent to the utility component in MAUT or AHP) and the weights are concurrently separated and measured as results of the examination. Such separation rendered measurement of the psychological weights with theoretical validity possible, and could be used as the standard so that weights measured with the other three methods are evaluated and compared.
Findings of the research are as follow, in the aspect of multiple-attribute weight measurement methods applied in the place marketing of Kaohsiung¡G
1. Among the three major factors, the most important is ¡§economic development¡¨, the second is ¡§government function¡¨, and then the third is ¡§social and political development¡¨.
2. For economic development, under the situation of maintain status quo, the most crucial derivative attribute is¡§ the level of economic globalization of Kaohsiung¡¨. The second one is ¡§the industrial structure of Kaohsiung city¡¨. The last one is ¡§the competitiveness of Kaohsiung Port and Airport project¡¨¡FUnder the situation of opening for three direct-links, the most crucial derivative attribute is still ¡§the level of economic globalization of Kaohsiung¡¨. The second one is ¡§the competitiveness of Kaohsiung Port and Airport project¡¨. The last one is ¡§the industrial structure of Kaohsiung city¡¨.
3. For the factor of social and political development, the most important attribute is ¡§the level of citizen participation in public affairs and the extent they influence public policies¡¨, the second is ¡§the public values and culture¡¨, the last is ¡§the operational mechanism of politics and election¡¨.
4. As for government function, the most essential attribute is ¡§the provision of public infrastructures¡¨. The second is ¡§the elected officials¡¦ perception of place marketing and capabilities of policy planning¡¨. The third is¡§the career officials¡¦ identifications for the elected officials¡¦ ambitions and abilities to carry out the place marketing policies¡¨.
5. The second group examinees think that direct-links has positive effects on the developments of Kaohsiung and most of them approve of the policy of three direct-links¡Fthe first group examinees have diverse opinions on it.
6. Based on PAM framework, we will sum up the problems of place marketing of Kaohsiung and brings out policy suggestions for Kaohsiung city.
In the aspect of Multiple-attribute weight measurement methods theory¡G
1. Take the weight parameters derived from IIT¡¦s averaging model as the standard, weights from rating¡BAHP and SMART are all theoretically valid. The theoretical validity of rating outperforms that of AHP, AHP outperforms that of SMART. Among the convergence validity or consistence of four methods, between rating and AHP with IIT¡¦s averaging model fit better in convergence validity, between SMART and IIT¡¦s averaging model don¡¦t.
2. All three models had fairly high predictability and notable for subjects¡¦ holistic judgments. The predictability of IIT is the a little bit higher than SMART, the third one is AHP, and the last one is rating.
3. For convergence, measured with accumulative weights derived from subjects selected 15-20 examinees randomly, IIT was better than SMART, and SMART was better than AHP.
4. In terms of perceived performance, degree of operational easiness and fondness, the three models are not significantly different. As for the degree of confidence, IIT outperformed AHP, and AHP outperformed SMART.
|
Page generated in 0.0597 seconds