다중혈관 관상동맥 환자에서 y-문합을 이용하여 양쪽 내흉동맥만을 사용한 우회술의 조기 성적

Clustering algorithms are attractive for the task of class identification in spatial databases. However, the application to large spatial databases rises the following requirements for clustering algorithms: minimal requirements of domain knowledge to determine the input parameters, discovery of clusters with arbitrary shape and good efficiency on large databases. The well-known clustering algorithms offer no solution to the combination of these requirements. In this paper, we present the new clustering algorithm DBSCAN relying on a density-based notion of clusters which is designed to discover clusters of arbitrary shape. DBSCAN requires only one input parameter and supports the user in determining an appropriate value for it. We performed an experimental evaluation of the effectiveness and efficiency of DBSCAN using synthetic data and real data of the SEQUOIA 2000 benchmark. The results of our experiments demonstrate that (1) DBSCAN is significantly more effective in discovering clusters of arbitrary shape than the well-known algorithm CLAR-ANS, and that (2) DBSCAN outperforms CLARANS by a factor of more than 100 in terms of efficiency.

A density-based algorithm for discovering clusters a density-based algorithm for discovering clusters in large spatial databases with noise

Clustering algorithms are attractive for the task of class identification in spatial databases. However, the application to large spatial databases rises the following requirements for clustering algorithms: minimal requirements of domain knowledge to determine the input parameters, discovery of clusters with arbitrary shape and good efficiency on large databases. The well-known clustering algorithms offer no solution to the combination of these requirements. In this paper, we present the new clustering algorithm DBSCAN relying on a density-based notion of clusters which is designed to discover clusters of arbitrary shape. DBSCAN requires only one input parameter and supports the user in determining an appropriate value for it. We performed an experimental evaluation of the effectiveness and efficiency of DBSCAN using synthetic data and real data of the SEQUOIA 2000 benchmark. The results of our experiments demonstrate that (1) DBSCAN is significantly more effective in discovering clusters of arbitrary shape than the well-known algorithm CLARANS, and that (2) DBSCAN outperforms CLARANS by a factor of more than 100 in terms of efficiency.

/pdf/a-density-based-algorithm-for-discovering-clusters-in-large-1o5u8sf38m.pdf

A density-based algorithm for discovering clusters in large spatial Databases with Noise

[신간의 별자리x] 우리/미술, 그리고 ‘슬픔의 박물관’

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Bayesian methods and spatial stochastic processes are used in the deconvolution of images of galaxies. Under very simple but realistic prior assumptions about the true underlying image of a galaxy the Bayesian framework is put to work. The method is tested in CCD images of extragalactic objects of different morphological types and an analysis of the deconvolutions obtained is performed emphazing the comparison with other observational results

Bayesian Deconvolution in Optical Astronomy

Background The diversity of the Mediterranean diet and the heterogeneity of acquired epigenetic alterations in colorectal cancer (CRC) led us to examine the possible association between dietary factors and promoter hypermethylation in genes implicated in the pathogenesis of these neoplasms (p16(INK4a), p14(ARF), hMLH1) and the interaction with methylene tetrahydrofolate reductase (MTHFR) C677T polymorphism. Patients and methods For the molecular study, 120 CRC patients were analyzed for hMLH1 promoter methylation status and MTHFR genotyping. Dietary patterns and molecular data on p16(INK4a) and p14(ARF) methylation were obtained from previous studies with this populations. Results Patients with methylation in p16(INK4a) consumed significantly less folate (p = 0.01), vitamin A (p = 0.01), vitamin B1 (p = 0.007), potassium (p = 0.03) and iron (p = 0.02) than controls. Patients with methylation in p14(ARF) or hMLH1 consumed significantly less vitamin A (p = 0.001 and p = 0.05, respectively). Conclusion These results support that certain micronutrients protect against colorectal neoplasia and emphasize the importance of considering the different molecular forms of CRC as etiologically distinct diseases.

Lower specific micronutrient intake in colorectal cancer patients with tumors presenting promoter hypermethylation in p16(INK4a), p4(ARF) and hMLH1.

A number of restoration filters have been proposed for the restoration problem from partially-known blurs. Previously we proposed the regularized constrained least-squares filter (RCTLS) and showed that it has a number of advantages over previous ones (Mesarovic et al. 1995). However, the problem of estimating the parameters that define the RCTLS filter has not yet been addressed. In this paper we propose a two-step algorithm based on the hierarchical Bayesian approach to simultaneously restore the image and estimate the parameters of the RCTLS restoration filter. The algorithm is derived in the DFT domain; thus, it is very efficient even for very large images.

/pdf/hierarchical-bayesian-image-restoration-from-partially-known-1xns9ik559.pdf

Hierarchical Bayesian image restoration from partially-known blurs

BACKGROUND The deleted GSTT1 and GSTM1 genotypes (null genotypes) resulting in loss of transferase activity are found in 10-20% and 50-60% of the population, respectively. PATIENTS AND METHODS The GSTT1- and GSTM1-dependent risk for sporadic colorectal cancer (CRC) was studied in 247 incident CRC cases and 296 hospital-based controls. RESULTS The GSTT1-null genotype was found to be 1.5 times more prevalent in CRC patients (17.4%) compared with controls (11.1%) (crude OR 1.6; p = 0.03). The GSTM1-null genotype was found to be equally prevalent in cases and controls (53%). Multivariate analysis showed a significant 1.7-fold risk for CRC associated with the GSTT1-null genotypes (p = 0.04) and this increased to 2.9 for smokers (p = 0.02). CONCLUSION This study provides evidence of gene-environment interaction and illustrates the importance of further research into the role of genetic susceptibility for CRC.

Glutathione S-transferase (GSTM1 and GSTT1)-dependent risk for colorectal cancer.

There has been a significant interest in the recovery of low-rank matrices from an incomplete of measurements, due to both theoretical and practical developments demonstrating the wide applicability of the problem. A number of methods have been developed for this recovery problem, however, a principled method for choosing the unknown target rank is generally missing. In this paper, we present a recovery algorithm based on sparse Bayesian learning (SBL) and automatic relevance determination principles. Starting from a matrix factorization formulation and enforcing the low-rank constraint in the estimates as a sparsity constraint, we develop an approach that is very effective in determining the correct rank while providing high recovery performance. We provide empirical results and comparisons with current state-of-the-art methods that illustrate the potential of this approach.

/pdf/low-rank-matrix-completion-by-variational-sparse-bayesian-2su0fj8abp.pdf

Rafael Molina

Papers

Bayesian Deconvolution in Optical Astronomy

Lower specific micronutrient intake in colorectal cancer patients with tumors presenting promoter hypermethylation in p16(INK4a), p4(ARF) and hMLH1.

Hierarchical Bayesian image restoration from partially-known blurs

Glutathione S-transferase (GSTM1 and GSTT1)-dependent risk for colorectal cancer.

Low-rank matrix completion by variational sparse Bayesian learning