Thesis (M. Phil.)--Hong Kong University of Science and Technology, 2003. / Includes bibliographical references (leaves 62-65). Also available in electronic version. Access restricted to campus users.

Underling neuronal loss and subsequent hippocampal sclerosis, as reflected by hippocampal atrophy on structural magnetic resonance (MR) imaging, are the dominant findings in the patient with mesial temporal lobe epilepsy (MTLE). Yet, prolongation of T2 relaxation time has also been reported as an early marker for MTLE, but it is a rather insensitive marker. Typical age-related atrophy often constitutes a significant confounding factor, and atrophy often represents a late sign in hippocampal sclerosis. In this connection, there is an urge for a sensitive independent predictor for the early detection of MTLE. T1rho MR imaging provides a distinct contrast mechanism in tissue characteristics. It is sensitive to physio-chemical processes and has been tested successfully in Alzheimer’s disease, Parkinson’s disease and certain brain tumors. Therefore, it is possible to depict early biochemical change in patients with MTLE by means of measuring the changes in T1rho relaxation time. T1rho relaxation time is not affected by age-related atrophic changes and thus can be used as an independent marker. In this preliminary study, we aimed to assess the feasibility of T2 relaxometry and T1rho MR imaging in identification of the atrophied zones in patients with MTLE. Seven patients with unilateral MTLE and fourteen normal subjects were recruited. Three-dimensional T1-weighted imaging, axial T2 relaxometry and T1rho imaging were performed on a 3T MR scanner. Hippocampal head, hippocampal body, hippocampal tail and amygdala were contoured on the axial T2-weighted images and then co-registered onto T2 relaxometry and T1rho images. A combination of visual and quantitative volumetric assessment was used as the primary end outcome. For T2 relaxometry and T1rho imaging, their respective relaxation times together with the corresponding right-left asymmetric ratios were calculated for subsequent analysis. Abnormal right-left asymmetric ratio is defined as a deviation of 2SD from the mean of the Z-score. In the lateralizing epileptogenic zones, T1rho yielded an overall accuracy of 92.9% (sensitivity 100%, specificity 60%), while T2 relaxometry yielded an overall accuracy of 71.4% (sensitivity 65.2%, specificity 100%) only. T1rho imaging is thus superior to T2 relaxometry (P = 0.036, by chi-square test). To conclude, the present study indicated that T1rho is feasible and potentially useful to serve as a non-invasive imaging tool in the detection of lateralization of the epileptogenic zone in patients with MTLE. It can also facilitate prompt diagnosis and longitudinal disease monitoring. In addition, the generation of associated color-coded parametric map can provide an easy mean for direct visual analysis. / published_or_final_version / Diagnostic Radiology / Master / Master of Philosophy

Not available / text

By improving the precision and accuracy of public health surveillance tools, we can improve cost-efficacy and obtain meaningful information to act upon. In this dissertation, we propose statistical methods for improving public health surveillance research. In Chapter 1, we introduce a pooled testing option for HIV prevalence estimation surveys to increase testing consent rates and subsequently decrease non-response bias. Pooled testing is less certain than individual testing, but, if more people to submit to testing, then it should reduce the potential for non-response bias. In Chapter 2, we illustrate technical issues in the design of neonatal tetanus elimination surveys. We address identifying the target population; using binary classification via lot quality assurance sampling (LQAS); and adjusting the design for the sensitivity of the survey instrument. In Chapter 3, we extend LQAS survey designs for monitoring malnutrition for longitudinal surveillance programs. By combining historical information with data from previous surveys, we detect spikes in malnutrition rates. Using this framework, we detect rises in malnutrition prevalence in longitudinal programs in Kenya and the Sudan. In Chapter 4, we develop a computationally efficient geostatistical disease mapping model that naturally handles model fitting issues due to temporal boundary misalignment by assuming that an underlying continuous risk surface induces spatial correlation between areas. We apply our method to assess socioeconomic trends in breast cancer incidence in Los Angeles between 1990 and 2000. In Chapter 5, we develop a statistical framework for addressing statistical uncertainty associated with denominator interpolation and with temporal misalignment in disease mapping studies. We propose methods for assessing the impact of the uncertainty in these predictions on health effects analyses. Then, we construct a general framework for spatial misalignment in regression.

Conditional Random Fields (CRFs) is a discriminative and supervised approach for simultaneous sequence segmentation and frame labeling. Latent-Dynamic Conditional Random Fields (LDCRFs) incorporates hidden state variables within CRFs which model sub-structure motion patterns and dynamics between labels. Motivated by the success of LDCRFs in gesture recognition, we propose a framework for automatic facial expression recognition from continuous video sequence by modeling temporal variations within shapes using LDCRFs. We show that the proposed approach outperforms CRFs for recognizing facial expressions. Using Principal Component Analysis (PCA) we study the separability of various expression classes in lower dimension projected spaces. By comparing the performance of CRFs and LDCRFs against that of Support Vector Machines (SVMs) and a template based approach, we demonstrate that temporal variations within shapes are crucial in classifying expressions especially for those with small facial motion like anger and sadness. We also show empirically that only using changes in facial appearance over time without using the shape variations fails to obtain high performance for facial expression recognition. This reflects the importance of geometric deformations on face for recognizing expressions. / text

No description available.

No description available.

No description available.

Memory for visual patterns was examined in 131 patients with unilateral temporal- or frontal-lobe excisions and 32 normal control subjects. A deficit in short-term memory for matrices of increasing complexity was exhibited by the right frontal-lobe group. Right temporal lobectomy impaired cued recall of visually homogeneous matrices at each of four serial positions. On the delayed recognition of complex visual scenes, right temporal lobectomy decreased identification of changes in figurative detail and spatial composition, whereas right hippocampectomy impaired identification of changes in spatial location. The interplay between verbal and pictorial codes in memory was also investigated using related word-design pairs. Right or left temporal lobectomy affected the number of designs recalled but only the right temporal-lobe group produced designs of poor quality. When cued with the words, the left temporal-lobe group produced fewer designs than the control subjects, demonstrating a reduced ability to retrieve pictorial information through verbal labels.

No description available.