[Truncated abstract] Background: Dementia, largely due to Alzheimer's disease (AD), is a major public health problem. The early identification of disease is an important challenge for clinicians because treatment of AD is now available. A simple and accurate means of stratifying risk for AD and identifying early disease is needed so that risk factor modification and treatment can occur optimally. To date, despite many attempts, an accurate means of standardising an approach to the assessment of subtle cognitive symptoms has not been developed. A subjective complaint of poor memory has been identified as a possible marker for underlying brain disease. This study examines the utility of neuropsychological scores, homocysteine levels, APOE genotyping and brain imaging as predictors of cognitive decline in individuals with subjective memory complaint (SMC). Method Eighty subjects with SMC were recruited from memory clinics and the community (MC: 1). Forty-two control subjects were also examined (MC: 0). CAMDEX was used to describe baseline clinical features. The CAMCOG was used as a global test of cognition and was administered annually for four years. At baseline, neuropsychological testing was administered. Cranial CT scanning, measurement of plasma homocysteine and APOE genotyping were completed. Categorical variables were analysed using chi-square according to Pearson's method. Continuous data was analysed using Student's t-tests and Mann-Whitney tests. A logistic regression model was used to identify independent contributors to the presence of memory complaint. Participants were then matched for age, gender and time to follow-up (up for three years) to determine longitudinal predictors of cognitive decline. ... Baseline CAMCOG scores were greater in the control group (MC:0 = 98.3 ? 2.8, MC:1 94.2 ? 5.5, Z ?4.46, p 0.000). There were no differences in neuropsychological scores, concentration of total plasma homocysteine, APOE genotype or brain scan measurements. Using the Wald stepwise selection method, logistic regression could not be established due to non-convergence regardless of whether or not the continuous variables were re-coded into dichotomous variables. A matching process that created 32 pairs of controls/subjects allowed follow-up analysis. The controls showed significant improvement with time on the CAMCOG unlike subjects (mean ? SD, controls 1.5 ?-3.0, Z - 2.61, p 0.01, subjects 0.2 ? 3.2, Z ? 0.24, p 0.81). The logistic regression analysis showed that group membership could not be defined by any single independent variable. When group membership was abandoned and those with stable scores were compared to those who declined no clear meaningful independent predictors of decline apart from age were identified. Conclusions: Methodological issues such as small sample size and inadequate follow up duration were identified that may have precluded identification of predictive factors for cognitive decline. The results indicate that complaints of memory problems are not associated with established risk factors for Alzheimer's disease and fail to predict objective cognitive decline over three years. Future studies should continue trying to identify robust predictors of cognitive decline in later life.

The general purpose of this study was to investigate the relationship between neuropsychological tests scores and perfusion deficits, based upon measures of regional cerebral blood flow (rCBF) taken from the single photon emission computed tomography (SPECT) scans of patients suffering from dementia of the Alzheimer type (DAT). The study was designed to determine if DAT patients categorized as having left hemisphere, right hemisphere, diffuse, or an absence of perfusion deficits, as measured by SPECT, would be accurately grouped into their respective categories, and if they would exhibit the corresponding neuropsychological deficiencies usually associated with lateral hemispheric asymmetries.Selected subjects were 80 right handed, DAT patients from the North Broward Medical Center - Memory Disorder Center, in Pompono Beach, Florida, who had been administered a neuropsychological test battery, and a SPECT scan.Through several ANOVA's that were calculated for each of the neuropsychological variables, it was concluded that DAT patients who suffered from perfusion deficits exhibited significantly lower levels of neuropsychological functioning than DAT patients without perfusion deficits.These analyses revealed significantly lower levels of neuropsychological performance in the perfusion deficit group on the combination of left hemisphere WAIS-R subtests (Information, Similarities & Vocabulary), WMS - Logical Story (p < .01), WRAT-R Reading, WRAT-R Mathematics, WMS Paired Associates, and the Rey Complex Figure (p < .05).A separate step-wise discriminant function analysis indicated that a combination of the neuropsychological variables could not accurately classify the DAT patients into their respective right hemisphere, left hemisphere, diffuse, or absence of perfusion deficit groups. The discriminant function classified only 32.5% of the grouped cases accurately. Of the original thirteen neuropsychological variables, only Paired Associates immediate recall of the WMS entered the discriminant analysis equation. This accounted for only 23% of the total variability that could be explained by differences between the perfusion deficit groups. In several post hoc ANOVA's using the Bonferroni method of multiple comparisons, it was revealed that the absence of perfusion deficit group scored significantly higher than the other groups on the majority of the left hemisphere neuropsychological measures. However, none of the right hemisphere neuropsychological measures attained significance. / Department of Educational Psychology

This study investigated the utility of the cognitive measures of the Dean-Woodcock Neuropsychological Assessment System (D-WNAS) in the differential diagnosis of Alzheimer dementia (AD) from depression. Past research has found an overlap of symptoms in the early stages of AD and those found in geriatric depression. In both instances, patients are likely to report memory loss, attention deficits, and mood disturbances. As a result of this similarity, differentially diagnosing one from another is a vexing problem for the clinical practitioner. Although a number of screening measures have been offered, none have proven to be clinically useful. Some have proposed this is the result of reliance upon use of single-factor measures. Indeed, many have proposed a multiple factor assessment model would be of more utility in diagnosing AD and depression. Considering the importance of an accurate diagnosis in treatment, this study utilized a multiple factor cognitive model offered by the Dean-Woodcock Neuropsychological Assessment System to differentiate AD from depression.Specifically, subtest scores of the Woodcock-Johnson III - Tests of Cognitive Ability (WJ-III; cognitive measure of the Dean-Woodcock Neuropsychological Assessment System) were compared. Participants (n = 172) fell into one of three groups (i.e. Depressed, Demented, or Normal) based on the diagnoses of a board certified neurologist and neuropsychologist. Results showed clinical groups performed more poorly than normal participants on tests of the WJ-III. In addition, AD participants differed significantly from depressed participants on the Visual Matching and Spatial Relations tests of the WJ-III. However, in all, the WJ-III demonstrated a classification hit rate of less than 70%. Although groups were found to differ in specific ways, the classification hit rate of the WJ-III suggested it could not differentially diagnose AD from depression alone. / Department of Educational Psychology

Doctor of Philosophy(PhD) / In recent years there has been a substantial increase in research examining the role of subjective memory complaints (SMC) in cognitive function and Alzheimer’s disease. These studies have related SMC to many different cognitive outcomes, such as retaining normal cognitive function, a fluctuating cognitive performance and the development of Alzheimer’s disease. Most of these studies have focused on older populations and have employed a limited assessment of cognitive function. This limits the available evidence regarding the clinical utility of SMC. The literature on the role of SMC in younger subjects is scarce. It is not known whether memory complaints are useful in predicting future cases of Alzheimer’s disease in younger community-based subjects. Aims: The main aim of the present study was to determine whether SMC predict the development of cognitive impairment in a younger cohort of subjects, many of whom were under the age of 70 years (73%), based on their risk profile and neuropsychological assessment. A further aim was to ascertain whether the DRS or 7MS are sensitive screening tools for MCI and examine whether the presence of SMC affects the 3-year cognitive outcome of subjects. To address these aims, this study consisted of two parts: a cross-sectional design and a longitudinal follow-up component. Methods: This study was carried out with 86 community-dwelling subjects recruited via advertisement within the catchment area of Central Sydney Area Health Service. The mean age of the subjects was 63.1 years (SD=8.4). Subjective memory complaints were assessed using a single question. Cognitive function was assessed using a comprehensive battery of tests, selected on the basis of their sensitivity to identifying cognitive impairment typically associated with Alzheimer’s disease. After the initial analysis between those with SMC and without SMC, subjects were further classified according to their performance on an episodic memory task (i.e., delayed verbal recall, Rey, 1964) as having normal memory function, SMC or aMCI. Results: Part 1 - Subjective memory complaints (SMC) were reported by 63% of the sample. The initial analysis between subjects with SMC (n=54) and without SMC (n=32) suggested an initial relationship between SMC and cognitive functioning. Subjects with SMC had impaired global cognitive functioning on two brief screening tests (7MS and DRS), working memory, verbal recall and visuomotor speed. However, subsequent screening with the delayed verbal recall test showed that 12 of the 54 subjects with SMC demonstrated significant cognitive impairment, scoring 2 SD below the control group mean. After these subjects were removed to form the aMCI group, the cognitive differences between subjects with SMC and without SMC were no longer apparent. Subjects with aMCI showed evidence of multiple cognitive deficits (below 1 SD of control group mean) with a high percentage of subjects demonstrating impairment on tests of verbal learning, verbal recall, verbal ability and visuomotor speed. Further analysis showed a significant association between age and subjects identified as having SMC (r=-.581, p<.001) and aMCI (r=.692, p<.001). From the age of 60 onwards, both the SMC and aMCI groups demonstrated a more rapid cognitive decline with increasing age in several cognitive domains. Part 2 - After a mean interval of 3.2 years, 43 subjects were followed up. Subjects with aMCI showed evidence of greater decline on both screening tests (7MS; DRS), whilst the SMC group had significantly higher scores. This trend was also apparent with other neuropsychological testing. The analysis of change over time in cognitive function showed that the majority of subjects (both SMC aMCI) either remained stable or improved their cognitive performance. It is likely that the small sample size and short follow-up interval of the present study contributed to the present observation of no change in cognitive function over time. Discussion: The present findings suggest that subjective memory complaints are a poor predictor of cognitive function. In isolation, SMC are unlikely to be useful for identifying cases with significant cognitive impairment. This is particularly relevant for subjects under the age of 70 years. However, for subjects over the age of 70 years, SMC are likely to identify significant cases with neuropsychological assessment (such as animal fluency and delayed recall). Conclusion: The present study showed that SMC are a poor predictor of cognitive function in subjects under the age of 70 years. This study provided evidence that selected and relatively quick to administer formal neuropsychological tests of cognitive function (in particular tests of animal fluency and delayed recall) are better able to identify those at risk of developing cognitive impairment associated with Alzheimer’s disease, at an earlier age. This would thus allow exposure to earlier treatment options, such as donepezil, aricept, vitamin E, and memantine”.

Doctor of Philosophy(PhD) / In recent years there has been a substantial increase in research examining the role of subjective memory complaints (SMC) in cognitive function and Alzheimer’s disease. These studies have related SMC to many different cognitive outcomes, such as retaining normal cognitive function, a fluctuating cognitive performance and the development of Alzheimer’s disease. Most of these studies have focused on older populations and have employed a limited assessment of cognitive function. This limits the available evidence regarding the clinical utility of SMC. The literature on the role of SMC in younger subjects is scarce. It is not known whether memory complaints are useful in predicting future cases of Alzheimer’s disease in younger community-based subjects. Aims: The main aim of the present study was to determine whether SMC predict the development of cognitive impairment in a younger cohort of subjects, many of whom were under the age of 70 years (73%), based on their risk profile and neuropsychological assessment. A further aim was to ascertain whether the DRS or 7MS are sensitive screening tools for MCI and examine whether the presence of SMC affects the 3-year cognitive outcome of subjects. To address these aims, this study consisted of two parts: a cross-sectional design and a longitudinal follow-up component. Methods: This study was carried out with 86 community-dwelling subjects recruited via advertisement within the catchment area of Central Sydney Area Health Service. The mean age of the subjects was 63.1 years (SD=8.4). Subjective memory complaints were assessed using a single question. Cognitive function was assessed using a comprehensive battery of tests, selected on the basis of their sensitivity to identifying cognitive impairment typically associated with Alzheimer’s disease. After the initial analysis between those with SMC and without SMC, subjects were further classified according to their performance on an episodic memory task (i.e., delayed verbal recall, Rey, 1964) as having normal memory function, SMC or aMCI. Results: Part 1 - Subjective memory complaints (SMC) were reported by 63% of the sample. The initial analysis between subjects with SMC (n=54) and without SMC (n=32) suggested an initial relationship between SMC and cognitive functioning. Subjects with SMC had impaired global cognitive functioning on two brief screening tests (7MS and DRS), working memory, verbal recall and visuomotor speed. However, subsequent screening with the delayed verbal recall test showed that 12 of the 54 subjects with SMC demonstrated significant cognitive impairment, scoring 2 SD below the control group mean. After these subjects were removed to form the aMCI group, the cognitive differences between subjects with SMC and without SMC were no longer apparent. Subjects with aMCI showed evidence of multiple cognitive deficits (below 1 SD of control group mean) with a high percentage of subjects demonstrating impairment on tests of verbal learning, verbal recall, verbal ability and visuomotor speed. Further analysis showed a significant association between age and subjects identified as having SMC (r=-.581, p<.001) and aMCI (r=.692, p<.001). From the age of 60 onwards, both the SMC and aMCI groups demonstrated a more rapid cognitive decline with increasing age in several cognitive domains. Part 2 - After a mean interval of 3.2 years, 43 subjects were followed up. Subjects with aMCI showed evidence of greater decline on both screening tests (7MS; DRS), whilst the SMC group had significantly higher scores. This trend was also apparent with other neuropsychological testing. The analysis of change over time in cognitive function showed that the majority of subjects (both SMC aMCI) either remained stable or improved their cognitive performance. It is likely that the small sample size and short follow-up interval of the present study contributed to the present observation of no change in cognitive function over time. Discussion: The present findings suggest that subjective memory complaints are a poor predictor of cognitive function. In isolation, SMC are unlikely to be useful for identifying cases with significant cognitive impairment. This is particularly relevant for subjects under the age of 70 years. However, for subjects over the age of 70 years, SMC are likely to identify significant cases with neuropsychological assessment (such as animal fluency and delayed recall). Conclusion: The present study showed that SMC are a poor predictor of cognitive function in subjects under the age of 70 years. This study provided evidence that selected and relatively quick to administer formal neuropsychological tests of cognitive function (in particular tests of animal fluency and delayed recall) are better able to identify those at risk of developing cognitive impairment associated with Alzheimer’s disease, at an earlier age. This would thus allow exposure to earlier treatment options, such as donepezil, aricept, vitamin E, and memantine”.

Diabetes is a disease with a deleterious pathology that currently impacts 4.5 million individuals within the United States. This study examined the ability of a specific neuropsychological battery to identify and classify dementia type, investigated the impact of diabetes on cognition and analyzed the ability of the memory measures of the 7 Minute Screen (7MS) and the Rey-Osterrieth Recall to correctly categorize dementia type when not used in combination with a full battery. The battery in addition to exhaustive patient history, medical chart review and pertinent tests were used in initial diagnosis. Results indicated the battery was sufficient in the identification and classification of dementia type. Within the sample, diabetes did not appear to significantly impact overall battery results whereby only two measures were minimally affected by diabetes. Finally, the memory measures of the 7MS and the Rey-Osterrieth Recall were sufficient to predict membership into the Alzheimer's (AD) and vascular dementia (VD) groups with 86.4% accuracy. The classification percentage dropped to 68.3% with addition of the mild cognitive impairment category. The full battery correctly classified AD and VD dementia 87.5% and appeared to be the most robust.

Alzheimer’s disease and related dementias (ADRD) are among the leading causes of disability and mortality among the older population worldwide and a costly public health issue, yet there is still no treatment for prevention or cure. Clinical trials are available, but successful recruitment has been a longstanding challenge. One strategy to improve recruitment is conducting eligibility prescreening, a resource-intensive process where clinical research staff manually go through electronic health records to identify potentially eligible patients. Natural language processing (NLP), an informatics approach used to extract relevant data from various structured and unstructured data types, may improve eligibility prescreening for ADRD clinical trials. Guided by the Fit between Individuals, Task, and Technology framework, this dissertation research aims to optimize eligibility prescreening for ADRD clinical research by evaluating the sociotechnical factors influencing the adoption of NLP-driven tools. A systematic review of the literature was done to identify NLP systems that have been used for eligibility prescreening in clinical research. Following this, three NLP-driven tools were evaluated in ADRD clinical research eligibility prescreening: Criteria2Query, i2b2, and Leaf. We conducted an iterative mixed-methods usability evaluation with twenty clinical research staff using a cognitive walkthrough with a think-aloud protocol, Post-Study System Usability Questionnaire, and a directed deductive content analysis. Moreover, we conducted a cognitive task analysis with sixty clinical research staff to assess the impact of cognitive complexity on the usability of NLP systems and identify the sociotechnical gaps and cognitive support needed in using NLP systems for ADRD clinical research eligibility prescreening. The results show that understanding the role of NLP systems in improving eligibility prescreening is critical to the advancement of clinical research recruitment. All three systems are generally usable and accepted by a group of clinical research staff. The cognitive walkthrough and a think-aloud protocol informed iterative system refinement, resulting in high system usability. Cognitive complexity has no significant effect on system usability; however, the system, order of evaluation, job position, and computer literacy are associated with system usability. Key recommendations for system development and implementation include improving system intuitiveness and overall user experience through comprehensive consideration of user needs and task completion requirements; and implementing a focused training on database query to improve clinical research staff’s aptitude in eligibility prescreening and advance workforce competency. Finally, this study contributes to our understanding of the conduct of electronic eligibility prescreening for ADRD clinical research by clinical research staff. Findings from this study highlighted the importance of leveraging human-computer collaboration in conducting eligibility prescreening using NLP-driven tools, which provide an opportunity to identify and enroll participants of diverse backgrounds who are eligible for ADRD clinical research and accelerate treatment development.

Accurate, early diagnosis of Alzheimer's Disease is becoming increasingly important in light of its growing prevalence among the expanding older-aged adult population. Due to its ability to assess multiple domains of cognitive functioning and provide a profile of impairment rather than a simple global score, the Neurobehavioral Cognitive Status Examination (NCSE) is suggested to better assess such patterns of cognitive deficit for the purpose of diagnosis. The performance of the NCSE was compared with that of the Mini-Mental State Examination (MMSE) for diagnostic sensitivity in a sample of patients diagnosed as having probable Alzheimer's Disease. The strength of correlation between severity of cognitive impairment on these tests and report of behavior problems on the Memory and Behavior Problems Checklist (MBPC) was also explored, as was performance on the NCSE and report of behavior problems using the MBPC in predicting Single Photon Emission Computed Tomography (SPECT) scan results. The NCSE was found to exhibit greater sensitivity to physician diagnosis of probable Alzheimer's Disease relative to two versions (Serial 7's or WORLD) of the MMSE (.90, .77 and .68, respectively). While both measures were found to correlate significantly with the report of behavior problems, only a moderate proportion (NCSE = .22 and MMSE = .33) of the explained variance was accounted for by either test. Severity of cognitive impairment on the NCSE was found to be significant, though small in estimate of its effect size, for predicting the absence/presence of pathognomic findings on SPECT scans. In contrast, the report of behavior problems on the MBPC did not significantly predict SPECT scan outcomes. The NCSE would appear to be a sensitive tool for the identification of the extent and severity of cognitive impairment found among demented individuals; however, it may be "over"-sensitive to such diagnosis. Although relationships between cognitive impairment and behavior problems and/or neuroradiological findings are observed, their meaningfulness remains with the need for further, more detailed, study using standardized criteria for comparison purposes.

Although not all persons with mild cognitive impairment (MCI) go on to develop Alzheimer's disease (AD), MCI is recognized as an early stage of AD. The effects of AD are devastating to all concerned. Research has identified that recognition of AD in its earliest stages and institution of known treatment modalities can forestall the ultimate outcome. Identification of the first subtle signs of MCI can assist in the recognition of this prodromal phase, and allow for institution of therapy while still in the initial stages. Unfortunately, the development of MCI is insidious in nature, thus making it difficult to detect. The purpose of this study was to identify areas of functional decline that occur in MCI in an effort to improve its early identification. A mixed-methods design that combined qualitative and quantitative methods was used. Fifty-three participants with memory complaints were interviewed using a semi structured interview technique with open-ended questions, the Montreal Cognitive Assessment (MoCA), the Geriatric Depression Scale (GDS) and a list of eighty-five items previously identified as indicative of functional decline. Twenty-nine persons were divided into two groups: 1) those identified as probable MCI (consensus diagnosis) (n=15) and possible MCI (based on screening examination) (n=14) and 2) those identified as Normal (no cognitive impairment) (n=10), and their subjective functional deficits compared. The findings suggest that there were certain areas of functional decline more commonly experienced by persons in the MCI group than by unimpaired. These include difficulty recalling details of information and forgetting conversations. There were also other changes identified, such as adaptations on the part of persons with MCI (an increased dependence on memory aids, for example, lists and calendars) and a dec rease in social activities leading to an increase in social isolation. Additionally identified were functional activities that appear to remain intact in persons with early MCI. This study highlights the subtlety with which MCI assaults the functional abilities of individuals, thus making its early identification problematic. The results of this study will contribute by providing information that will help professionals who are assessing persons experiencing memory issues for the possible presence of MCI. Additionally, it is hoped that these findings will assist in the development of a measurement tool designed to assess for possible MCI. / Includes bibliography. / Dissertation (Ph.D.)--Florida Atlantic University, 2015. / FAU Electronic Theses and Dissertations Collection

In conclusion, gene expression profiling in blood may have potential to be an adjuvant marker for early detection of AD. Expression marker panel is more informative than single gene expression signature. Further validation in prospective studies will substantiate its clinical application and explore its potential to differentiate AD from other dementias and to predict the progression from MCI to AD. (Abstract shortened by UMI.) / In the study, the profile of 12 target gene expression levels in peripheral blood cells were determined in 96 AD, 145 MCI and 167 normal controls (NC) by quantitative real-time RT-PCR. The genes were identified with (i) high expression in blood and brain; (ii) differential expression between AD and control; (iii) AD related candidate genes. Then, a list of genes were selected including CTSB, CTSD, DDT, ITPKB, NDUFA6, NRD1, PIN1, SNX2, TSC1, UQCRC1, CNR2, GSTM3. Seven genes were found to be differentially expressed between AD and NC group, with upregulation of CTSB, CTSD, DDT, TSC1 and UQCRC1, and downregulation of ITPKB and PIN1 in AD patients. Expression levels of two genes were increased in the MCI compared with NC group, including CTSB and CTSD. In addition, an upregulation of CTSD, UQCRC1, NRD1 and downregulation of ITPKB were observed in AD subjects in comparison to the MCI group (Mann-Whitney U test, p&lt;0.05). After adjusting for confounding factors of age, gender, education level, ApoE4 status and the total CIRS score, expression level of any single gene was not associated with the classfication of AD or MCI (Logistic regression, p>0.05). A five gene biomarker panel, including DDT, ITPKB, PIN1, TSC1 and UQCRC1 was identified with logistic regression analysis. The function of Logit(P)= ln(P/(1-P))= b0+b1RatioDDT+ b2RatioITPKB + b3Ratio PIN1 +b4 RatioTSC1+b5Ratio UQCRC1 were defined as the probability of a subject to be diagnosed as "AD" or "MCI' by using 5-gene biomarker panel. ROC analysis showed that the AUC for the 5-gene biomarkers panel in differentiating between AD and NC, between MCI and NC, between AD and MCI were 0.79 (95% CI, 0.72-0.86; p&lt;0.001), 0.61 (95% CI, 0.53-0.69; p=0.007) and 0.68 (95% CI, 0.60-0.76; p&lt;0.001) respectively. The 5-gene combination was found to discriminate AD subjects from normal controls with good sensitivity and specificity of 70.7% and 86.7% respectively at an optimal cut-off point of 0.486. Low sensitivity (42.4%) and acceptable specificity (76.2%) were observed at a cut-off threshold of 0.505 when differentiating MCI from NC subjects. Between AD and MCI subjects, gene combination showed a sensitivity of 61% and specificity of 73.7% at a cut-off value of 0.496. / Several genes including CTSD, DDT, NDUFA6, TSC1 and UQCRC1 were found in association with the cognitive and psychiatric symptoms, indicating the role of genetic factors in moderating the presence of cognitive and NP profiles in demented individuals. / The aim of the present study is to evaluate the gene expression profiling of peripheral leukocytes in Chinese subjects with Alzheimer's disease (AD) and explored its potential of clinical application. Behavioral phenotypes of cognitive performance and neuropsychiatric assessment were also investigated in association with gene expression in AD. Persons with mild cognitive impairment (MCI), as an at-risk state between normal aging and clinical dementia, was also assessed in consideration that the information may provide a better understanding of the mechanisms involved in clinical progression of AD. / The genes identified in this study were involved in processes implicated in neurodegneration, including protein isomerization (PIN1), calcium disequilibrium and mitochondria insufficiency (ITPKB and UQCRC1), increased inflammatory response (DDT), apoptosis (CTSB and CTSD) and neurogeneration (NRD1 and TSC1). / Fu, Yan. / Adviser: Chiu Wa Lam. / Source: Dissertation Abstracts International, Volume: 73-02, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2010. / Includes bibliographical references (leaves 132-168). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.