IGrC: Cognitive and Motor Changes During Symptoms Development in Parkinson's Disease Patients

摘要

Cognitive symptoms are characteristic for neurodegenerative disease: there are dominating in the Alzheimer's, but secondary in Parkinson's disease (PD). However, in PD motor symptoms (MS) are dominating and their characteristic helps neurologist to recognize the disease. There are a large number of data mining publications that analyzed MS in PD. Present study is related to the question if development of cognitive symptoms is related to motor symptoms or if they are two independent processes? We have responded to this problem with help of IGrC (intelligent granular computing) approach. We have put together eye movement, neurological and neuropsychological tests. Our study was dedicated to 47 Parkinson's disease patients in two sessions: S#l - without medications (Med_(OFF)) and S#2 after taking medications (Med_(ON))- There were two groups of patients: Gr1 (23 patients) less advanced and Gr2 more advanced PD. We have measured Grl in three visits every 6 months: Gr1 _(VIS1), Gr1_(VIS2), Gr1_(VIS3) . Gr2 (24 patients) has only one visit (no visit number). With rough set theory (RST) that belongs to IGrC we have found from Gr2 three different sets of rules: a) general rules (G_(RUL)) determined by all attributes; b) motor related rules (M_(RUL)) - motor attributes; c) cognitive rules (C_(RUL)determined by cognitive attributes. By applying these different sets of rules to different Grl visits we have found different set of symptoms developments. With G_(RUL) we have found for Gr1_(VIS1) accuracy = 0.682, for Gr1_(VIS2) acc. = 0.857, for Gr1_(VIS3) acc. = 0.875. With M_(RUL) we have found for Gr1_(VIS1) acc. = 0.80, for Gr1_(VIS2) acc. = 0.933, for Gr1_(VIS3) acc. = 1.0. With Crul we have found for Gr1_(VIS1) acc. = 0.50, for Gr1_(VIS2) acc. = 0.60, for Gr1_(VIS3) acc. = 0.636. Cognitive changes are independent from the motor symptoms development.