Patient stratification with competing risks by multivariate Fisher distance

Bacciu, Davide; Jarman, Ian H.; Etchells, Terence A.; Lisboa, Paulo J. G.

doi:10.1109/ijcnn.2009.5179077

Cited by 9 publications

(5 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The process is ended by reaching a point where no further changes in assignment to cluster are seen. By also repeating the whole process starting with a range of different numbers of initial cluster centres and values for starting clusters, the most robust clustering patterns with greatest differentiation between clusters were identified (Bacciu et al, 2009). The statistical analyses used SPSS (release 17.0.0; SPSS, 2008) and MATLAB (version 7.8.0 R2009a; MATLAB, 2009) 64bit for clustering.…”

Section: Methodsmentioning

confidence: 99%

Using clustering techniques to identify localities with multiple health and social needs

et al. 2012

View full text Add to dashboard Cite

Development of health promoting policies requires an understanding not just of the interplay between different measures of health but also their relationship with broader education, criminal justice and other social issues. Methods to better utilise multi-sectoral data to inform policy are needed. Applying clustering techniques to 30 health and social metrics we identify 5 distinct local authority types, with poor outcomes for the majority of metrics concentrated in the same cluster. Clusters were distinguished especially by levels of: child poverty; breastfeeding initiation; children's tooth decay; teenage pregnancy; healthy eating; mental illness; tuberculosis and smoking deaths. Membership of the worst cluster (C5) was focused in Northern England which contains 15.7% of authorities analysed (n=324), but 63.0% of those in C5. The concentration of challenges in certain areas creates disproportionate pressures that may exceed the cumulative effects of individual challenges. Such distinct health clusters also raise issues of transferability of effective policies between areas with different cluster membership.

show abstract

Section: Methodsmentioning

confidence: 99%

Using clustering techniques to identify localities with multiple health and social needs

et al. 2012

View full text Add to dashboard Cite

show abstract

“…This approach to robustness analysis is summarised elsewhere (Bacciu et al 2009). In principle, it may be applied to any choice of initialisation-dependent clustering method.…”

Section: Methodsmentioning

confidence: 99%

“…One method relies on mapping the landscape of cluster separation against stability, for many initializations of the k-means clustering, from which to identify optimal and reproducible cluster partitions; this is termed a separation concordance (SeCo) map (Bacciu et al 2009). The alternative benchmark method directly optimizes a penalised objective function using a competitive neural network with repetition suppression (CoRe) which dynamically clusters the data into a parsimonious representation that covers the data space (Bacciu and Starita 2008).…”

Section: Introductionmentioning

confidence: 99%

Clustering of protein expression data: a benchmark of statistical and neural approaches

et al. 2010

View full text Add to dashboard Cite

Clustering issues are fundamental to exploratory analysis of bioinformatics data. This process may follow algorithms that are reproducible but make assumptions about, for instance, the ability to estimate the global structure by successful local agglomeration or alternatively, they use pattern recognition methods that are sensitive to the initial conditions. This paper reviews two clustering methodologies and highlights the differences that result from the changes in data representation, applied to a protein expression data set for breast cancer (n = 1,076). The two clustering methodologies are a reproducible approach to model-free clustering and a probabilistic competitive neural network. The results from the two methods are compared with existing studies of the same data set, and the preferred clustering solutions are profiled for clinical interpretation

show abstract

“…Hierarchical algorithm can be further divided into bottom-up and top-down algorithms and partitioning clustering divided into k-mean and k-modes algorithms. The k-modes algorithm as an extension to k-means for categorical data, by replacing kmeans with k-modes, introduce a different dissimilarity measure and update the modes with a frequency based method [4,5,6]. In its basic form the clustering problem is defined as the problem of finding homogeneous groups of objects in a given dataset.…”

Section: Introductionmentioning

confidence: 99%

K-modes Clustering Algorithm for Categorical Data

Sharma¹,

Gaud²

2015

IJCA

View full text Add to dashboard Cite

Partitioning clustering is generally performed using K-modes cluster algorithms, which work well for large datasets. A Kmodes technique involve random chosen initial cluster centre (modes) as seed, which lead toward that problem clustering results be regularly reliant on the choice initial cluster centre and non-repeatable cluster structure may be obtain. K-Modes technique has been widely applied to categorical data a clustering in replace means through modes. The pervious algorithms select the attributes on frequency basis but not provided better result. Proposed algorithm select attributes on information gain basis which provide better result. Experimental results showing the proposed technique provided better accuracy.

show abstract

Patient stratification with competing risks by multivariate Fisher distance

Cited by 9 publications

References 13 publications

Using clustering techniques to identify localities with multiple health and social needs

Using clustering techniques to identify localities with multiple health and social needs

Clustering of protein expression data: a benchmark of statistical and neural approaches

K-modes Clustering Algorithm for Categorical Data

Contact Info

Product

Resources

About