HONcode

This website is certified by Health On the Net Foundation. Click to verify. This site complies with the HONcode standard for trust worthy health information:
verify here.


Search only trustworthy HONcode health websites:

Affiliations

REACTION is affilliated with these programs and organisations:

REACTION is delivering better and more efficient healthcare services in Europe, thus supporting the Commissions activities in ICT for Health: eHealth.



The REACTION platform allows for the creation of inclusive applications with accessibility for all. The project supports the Commissions campaign: eInclusion - be part of it!



We Fight Spam

Facts



The REACTION project is a 4-year project started in 2010. It is partly funded by the European Commission under the 7th Framework Programme in the area of Personal Health Systems under Grant Agreement no. 248590

 Impressum   Privacy

Events

<< September 2017 >>
Mo Tu We Th Fr Sa Su
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30  

No events.


Newsletter


Registered Members Only

Previous Newsletters


Read previous issues of our newsletter here:

April 2014
August 2013
June 2012
August 2011
April 2011
November 2010

Share this

Bookmark and Share

eHealth at Facebook

Join the discussion on EC eHealth at Facebook!

Downloads: Biomarker signature identification in “omics” data with multi-class outcome

Downloads Home > Journal papers > Biomarker signature identification in “omics” data with multi-class outcome

Biomarker signature identification in “omics” data with multi-class outcome


Abstract: Biomarker signature identification in “omics” data is a complex challenge that requires specialized feature selection
algorithms. The objective of these algorithms is to select the smallest set(s) of molecular quantities that are able to predict a given outcome (target) with maximal predictive performance. This task is even more challenging when the outcome comprises of
multiple classes; for example, one may be interested in identifying the genes whose expressions allow discrimination among different types of cancer (nominal outcome) or among different stages of the same cancer, e.g. Stage 1, 2, 3 and 4 of Lung Adenocarcinoma (ordinal outcome). In this work, we consider a particular type of successful feature selection methods, named constraint-based, local causal discovery algorithms. These algorithms depend on performing a series of conditional independence tests. We extend these algorithms for the analysis of problems with continuous predictors and multi-class outcomes, by developing and equipping them with an appropriate conditional independence test procedure for both nominal and ordinal multi-class targets. The test is based on multinomial logistic regression and employs the log-likelihood ratio test for model selection. We present a comparative, experimental evaluation on seven real-world, high-dimensional, gene-expression datasets. Within the scope of our analysis the results indicate that the new conditional independence test allows the identification of smaller and better performing signatures for multi-class outcome datasets, with respect to the current alternatives for performing the independence tests.
Calendar
Louise Birch Riley
14. January 2014 13:11
Downloads
Download 723

Download:
Download
Download Stats Downloads: 113
Downloaded: 143191
Most Downloaded: REACTION brochure [ 4880 ]
Most Recent: REACTION brochure [ 4880 ]