EVALUATION OF MUSCLE DISEASES USING ARTIFICIAL NEURAL-NETWORK ANALYSIS OF P-31 MR SPECTROSCOPY DATA

Dermatomyositis is an autoimmune disease characterized by an erythemat ous rash and severe muscle weakness. P-31 Magnetic resonance spectrosc opy (MRS) provides quantitative data for longitudinal monitoring of di sease status and responses to immunosuppressive therapy. A disease var iant, amyopathic dermatomyositis, presents with a typical rash but no clinical muscle weakness. However, metabolic abnormalities in the oxid ative capacity of muscles of amyopathic patients during exercise were detected with P-31 MRS. Because MRS provided the best quantitative dat a for evaluating dermatomyositis, the P-31 metabolic parameters derive d from the MR spectra were further processed using an artificial neura l network (XERION). The neural network analyses provided additional cl inical information from the weighted correlations of multiple P-31 par ameters, namely, inorganic phosphate, phosphocreatine, ATP, phosphodie sters, and selected ratios. This investigation analyzes the relative i mportance of the various metabolic parameters for accurate patient cha racterization and provides insights into the pathogenesis of the disea se.