[Ultrasound sonographic diagnosis of myocarditis using quantitative texture analysis of 2-dimensional echocardiography images].

In a prospective study we tested the possibility to differentiate myocardial tissue by means of the quantitative texture analysis of two-dimensional echocardiographic images. 12 patients with a bioptically ascertained myocarditis and 12 healthy patients of a control group were echocardiographically examined. The results demonstrated that the echocardiographic image texture in myocarditis is characterized by an increase of the texture parameters entropy, second difference moment and long-run emphasis as well as by a decrease of the parameters angular second moment, inverse difference moment and run-length nonuniformity as well as by different values of the ring and sector sums of the power spectre. The digital image processing of echocardiography images seems thus to open a new way to the noninvasive diagnostic of myocarditis.