Biomedical subjects

Shu-Kay Ng

Publications and source records attributed to Shu-Kay Ng.

4 recordsLinked to original sources

PubMed2006

Selenium binding protein 1 in ovarian cancer.

Selenium binding protein 1 (SELENBP1) was identified to be the most significantly down-regulated protein in ovarian cancer cells by a membrane proteome profiling analysis. SELENBP1 expression levels in 4 normal ovaries, 8 benign ovarian tumors, 12 borderline ovarian tumors and 141 invasive ovarian cancers were analyzed with immunohistochemical assay. SELENBP1 expression was reduced in 87% cases of invasive ovarian cancer (122/141) and was significantly reduced in borderline tumors and invasive cancers (p<0.001). Cox multivariate analysis within the 141 invasive cancer tissues showed that SELENBP1 expression score was a potential prognostic indicator for unfavorable prognosis of ovarian cancer (hazard ratio [HR], 2.18; 95% CI=1.22-3.90; p=0.009). Selenium can disrupt the androgen pathway, which has been implicated in modulating SELENBP1 expression. We investigated the effects of selenium and androgen on normal human ovarian surface epithelial (HOSE) cells and cancer cells. Interestingly, SELENBP1 mRNA and protein levels were reduced by androgen and elevated by selenium treatment in the normal HOSE cells, whereas reversed responses were observed in the ovarian cancer cell lines. These results suggest that changes of SELENBP1 expression in malignant ovarian cancer are an indicator of aberration of selenium/androgen pathways and may reveal prognostic information of ovarian cancer.

Aged↗

PubMed2005

An incremental EM-based learning approach for on-line prediction of hospital resource utilization.

OBJECTIVE: Inpatient length of stay (LOS) is an important measure of hospital activity, health care resource consumption, and patient acuity. This research work aims at developing an incremental expectation maximization (EM) based learning approach on mixture of experts (ME) system for on-line prediction of LOS. The use of a batch-mode learning process in most existing artificial neural networks to predict LOS is unrealistic, as the data become available over time and their pattern change dynamically. In contrast, an on-line process is capable of providing an output whenever a new datum becomes available. This on-the-spot information is therefore more useful and practical for making decisions, especially when one deals with a tremendous amount of data. METHODS AND MATERIAL: The proposed approach is illustrated using a real example of gastroenteritis LOS data. The data set was extracted from a retrospective cohort study on all infants born in 1995-1997 and their subsequent admissions for gastroenteritis. The total number of admissions in this data set was n = 692. Linked hospitalization records of the cohort were retrieved retrospectively to derive the outcome measure, patient demographics, and associated co-morbidities information. A comparative study of the incremental learning and the batch-mode learning algorithms is considered. The performances of the learning algorithms are compared based on the mean absolute difference (MAD) between the predictions and the actual LOS, and the proportion of predictions with MAD < or = 1 day (Prop(MAD < or = 1)). The significance of the comparison is assessed through a regression analysis. RESULTS: The incremental learning algorithm provides better on-line prediction of LOS when the system has gained sufficient training from more examples (MAD = 1.77 days and Prop(MAD < or = 1) = 54.3%), compared to that using the batch-mode learning. The regression analysis indicates a significant decrease of MAD (p-value = 0.063) and a significant (p-value = 0.044) increase of Prop(MAD < or = 1) with the incremental learning algorithm. CONCLUSIONS: The incremental learning feature and the self-adaptive model-selection ability of the ME network enhance its effective adaptation to non-stationary LOS data. It is demonstrated that the incremental learning algorithm outperforms the batch-mode algorithm in the on-line prediction of LOS.

Algorithms↗

PubMed2004

Using the EM algorithm to train neural networks: misconceptions and a new algorithm for multiclass classification.

The expectation-maximization (EM) algorithm has been of considerable interest in recent years as the basis for various algorithms in application areas of neural networks such as pattern recognition. However, there exists some misconceptions concerning its application to neural networks. In this paper, we clarify these misconceptions and consider how the EM algorithm can be adopted to train multilayer perceptron (MLP) and mixture of experts (ME) networks in applications to multiclass classification. We identify some situations where the application of the EM algorithm to train MLP networks may be of limited value and discuss some ways of handling the difficulties. For ME networks, it is reported in the literature that networks trained by the EM algorithm using iteratively reweighted least squares (IRLS) algorithm in the inner loop of the M-step, often performed poorly in multiclass classification. However, we found that the convergence of the IRLS algorithm is stable and that the log likelihood is monotonic increasing when a learning rate smaller than one is adopted. Also, we propose the use of an expectation-conditional maximization (ECM) algorithm to train ME networks. Its performance is demonstrated to be superior to the IRLS algorithm on some simulated and real data sets.

Algorithms↗

PubMed2002

Relationship of XIST expression and responses of ovarian cancer to chemotherapy.

Expression profiling to characterize cancer pharmacology has become a new approach to discover novel molecular targets for prognostic markers and cancer therapy. In a study to compare the global RNA expression profiles between primary and recurrent ovarian tumors from the same patient, we have identified XIST (inactive X chromosome-specific transcripts) as the most differentially expressed gene that was down-regulated in the recurrent tumor. XIST encodes a spliced noncoding polyadenylated transcript that is unique in being expressed exclusively from the inactive X chromosome and is involved in the X-inactivation process. Subsequent characterization of XIST expression in a panel of female cancer cell lines showed that the expression level of XIST correlates significantly with Taxol sensitivity. The clinical relevance of this observation is demonstrated by the strong association between XIST RNA levels and disease-free periods of ovarian cancer patients in a group of 21 ovarian cancer cases with Taxol in the therapeutic regiments. Cytogenetic studies on ovarian cancer cell lines indicated that loss of inactive X chromosome is one mechanism for the loss of XIST transcripts in the cell lines. Our data suggest that XIST expression may be a potential marker for chemotherapeutic responses in ovarian cancer.

Antineoplastic Agents, Phytogenic↗