Information Theoretic Quantification of Diagnostic Uncertainty

Abstract

Diagnostic test interpretation remains a challenge in clinical practice. Most physicians receive training in the use of Bayes’ rule, which specifies how the sensitivity and specificity of a test for a given disease combine with the pre-test probability to quantify the change in disease probability incurred by a new test result. However, multiple studies demonstrate physicians’ deficiencies in probabilistic reasoning, especially with unexpected test results. Information theory, a branch of probability theory dealing explicitly with the quantification of uncertainty, has been proposed as an alternative framework for diagnostic test interpretation, but is even less familiar to physicians. We have previously addressed one key challenge in the practical application of Bayes theorem: the handling of uncertainty in the critical first step of estimating the pre-test probability of disease. This essay aims to present the essential concepts of information theory to physicians in an accessible manner, and to extend previous work regarding uncertainty in pre-test probability estimation by placing this type of uncertainty within a principled information theoretic framework. We address several obstacles hindering physicians’ application of information theoretic concepts to diagnostic test interpretation. These include issues of terminology (mathematical meanings of certain information theoretic terms differ from clinical or common parlance) as well as the underlying mathematical assumptions. Finally, we illustrate how, in information theoretic terms, one can understand the effect on diagnostic uncertainty of considering ranges instead of simple point estimates of pre-test probability.

Information Gathering in Bayesian Networks Applied to Petroleum Prospecting

“…Shannon Entropy is probably the most used information criteria, and a few application areas are medicine in Westover et al (2012), piezometric data in Bueso et al (1998) and sulfate concentration records in Ko et al (1995). Fisher (1922) and Lindley (1956) both argue for their way of measuring information by ensuring that their measure has good properties.…”

Prevalence and factors predictive of intraocular fungal infection in patients with fungemia at an academic urban tertiary care center

“…Since the prevalence of disease affects pretest likelihood, a determination of the local prevalence of endogenous ocular fungal infections would be necessary to guide any changes to the screening algorithm of patients with fungemia. 16 …”