Not Logged In     Login  
 Release 22.0, Sept. 2008
 
Chapter : ch39. Decisional Analysis Section : Performance Measures
  Entropy of a Result Based on the Probability for a Disease

  Online Excel Reference
Copyright (c) 2008-2010, Institute for Algorithmic Medicine, Houston, TX, USA. All rights reserved.

Overview :

The entropy of a range of test results can provide insights into the test's usefulness to diagnose a binary outcome.

 

If a given finding is a associated with a probability of a binary outcome (presence or absence of a disease) then

 

entropy of the result = S =

= ((-1) * (probability) * LN(probability)) - ((1 - (probability)) * LN(1 - (probability)))

 

If this is rearranged:

 

(-1) * S =

= LN(((1 - prob)^(1 - (prob))) * ((prob)^(prob)))

 

The entropy for a result is maximal when the probability is 0.5 and minimal when either the disease has been excluded (probability 0) or is certain (probability 1).

 

Vollmer introduced the value (1 - S) which can be plotted vs a laboratory result. The lower the value (the higher the entropy) the lower the value of the information. A graphical plot of (1-S) against a range of test values can identify ranges with limited or maximal diagnostic utility.

 

  References:

Giasu S. Information Theory wit Applications. McGraw-Hill, 1977.

Vollmer RT. Entropy and information conotent of laboratory test results. Am J Clin Pathol. 2007; 127: 60-65.

 

 

   Online Excel | Reference TOP
     Pubmed Search For
Copy and paste the article title, or authors names into the search box