Statistical Decision In Medical Field, Two Examples

Two examples of statistical decision as applied to medical diagnosis are presented. The first one concerns the diagnosis of hypertension, i.e. essential hypertension and five types of secondary hypertension: fibrodysplasic renal artery stenosis, atheromatous renal artery stenosis, Conn's syndrome, renal cystic disease and pheochromocytoma. The second deals with the diagnosis of multiple sclerosis by visual, somatosensory and brain stem auditory evoked potentials. A set of discriminative and independant items is first selected by statistical investigation of experimental data (marginal d ensity distributions o f every item a nd correlation c oefficients). Joint density distribution functions are then established within every state. The expected loss corresponding to six possible decisions is estimated in the case of hypertension, wheras only the posterior bayesian probabilities are calculated for multiple sclerosis. The differential diagnosis thus established proves lo