Advertisements
47. Bayesian Confidence Propagation Neural Network

47. Bayesian Confidence Propagation Neural Network

The BCPNN (Bayesian Confidence Propagation Neural Network) is now routinely used in signal detection to search single drug – single ADR combinations. BCPNN not just aids in the early detection of adverse drug reactions (ADRs) but in addition further evaluation of such signals. BCPNN can also be used for detecting relation between group of similar drugs and particular ADR. 

BCPNN is a computational framework based on a statistical neural network where learning and inference is done using principles of Bayes law. 

The unique feature of this technique is that it considers all drug-ADR combinations in the database in an unbiased manner and it can early detect significant associations. It can calculate the inference despite of missing data with decreased certainty in the results. It is capable of performing the multi-variable analysis. 

In simple, BCPNN is an automated way of finding new drug-ADR combinations suspected to be signals through quantitative filtering of the database. This focuses clinical review on the potentially most important combinations of drugs and adverse reactions. 

This approach, when routinely applied to drug and adverse reaction combinations where variable x is the drug and variable y is the adverse reaction, can be seen as the calculation of the logarithm of the ratio of observed rate of adverse drug reactions to expected rate of adverse drug reactions under the null hypothesis of no association between drug and adverse reaction. The calculation is, however, done in a bayesian statistical framework.

The BCPNN uses Bayesian statistical methods implemented in the neural network architecture and the measure of disproportionality used, the Information Component (IC), is a Bayesian implementation of the observed to expected ratio, expressed as a base two logarithm. 

From IC probability distribution, expectation and variance values are calculated using Bayesian statistics. Thus estimates of precision (standard deviation) are provided for each point estimate of the IC, slowing both the point estimate and associated uncertainty to be examined. If a positive IC value increases over time and confidence interval narrows, this indicates an increased certainty of a positive quantitative association between the studies variables. 

The BCPNN method has been shown and tested for use in routine signal detection, refining signals and in finding complex patterns. The usefulness of the output is influenced by the quality of the data in the database. Therefore, this method should be used to detect, rather than evaluate signals. The need for clinical analyses of case series remains crucia

The detailed statistical application of BCPNN was described in article “A Bayesian recurrent neural network for unsupervised pattern recognition in large incomplete data sets

Advertisements

3 Comments

Leave a Reply

error: Content is protected !!