Development of a method for early and differential diagnosis of Parkinson's disease and essential tremor based on analysis of wave train electrical activity of muscles

The problems associated with the development of a method for early and differential diagnosis of Parkinson's disease (PD) and essential tremor (ET) using the analysis of wave train muscle activity are investigated. In the study of patients with PD, it is very difficult to find untreated patients who have not previously taken antiparkinsonian drugs. To solve the problem of small samples, it was proposed to increase artificially the sample of subjects using the bootstrap method. Three artificially enlarged samples of subjects were created: a sample of patients with PD, a sample of patients with ET, and a sample of control subjects. When constructing models, two problems arise: the problem of inadequacy and the problem of insufficient generalizing ability of the model. Both of these issues must be considered when building models. The paper proposes models that, according to the authors, are promising for the early and differential diagnosis of PD and ET. Using these models, a good accuracy of PD recognition at the early stage was obtained (about 100%); the accuracy of differential diagnostics (distinguishing between PD and ET) was more than 90%.