The potential use of biochemical-physiological simulation models in clinical chemistry

Publisher Website
Google Scholar
Abstract
The essence of the subject clinical chemistry is to describe, in chemical and biochemical terms, the different pathological processes causing illness. In the clinical chemistry laboratories billions of samples are processed to produce values for quantities related to components in different human body fluids or tissues. The directly obtained values of mass or substance amount concentrations are used to detect and characterize the severity of pathological processes. Such snapshots lead to static thinking rather than emphasizing the dynamics of such processes. If concentration values were used to derive values of other quantities, more closely describing the temporal progress of the pathological process, such transformed clinical chemical quantities might be more meaningful and more discriminative. A number of simple transformations of algebraic type are commonly used in clinical chemistry, some of which are purely empirical and some related more or less to basic processes. Combination of clinical chemical data by cluster analysis, discriminant function analysis, Bayesian probability techniques, etc., are methods requiring more calculations, but they are derived only from a purely statistical basis. More complex transformations, based on existing detailed biochemical and physiological knowledge, could be expected to become more frequently used in the future for transformation of concentration values into one or a few descriptive parameters more closely linked to pathological processes. How such transformations of the data should be achieved is discussed. The 1st step must be to describe pathological and relevant normal processes in the form of a conceptual model. Computer modeling and simulation are discussed.

This publication has 11 references indexed in Scilit: