Parkinson's Disease (PD) is a disorder in human movement coordination system that is characterized by motoric and non-motoric symptoms. At the late stage of PD, clinical diagnosis is relatively easy to detect because the symptoms are clear-cut. However, when the symptoms are often incomplete or subtle, in the initial stage, diagnosis becomes difficult and sometimes the subject still remains undiagnosed or even misdiagnosed. This study was aimed at identifying risk factors in early stage PD based on patient daily activities and rapid eye movement behaviour disorder. Daily activities were measured using the Modified Unified Parkinson's Disease Rating Scale (MDS-UPDRS) part I and part II and sleep behaviour disorder was measured using Rapid eye movement sleeping Behaviour Disorder Screening Questionnaire (RBDSQ). Data analysis was conducted using classification trees with CART algorithm, to classify patients into early stage PD patients or healthy control patients. Missing values were handled using k-Nearest Neighbour (kNN) method. The results were satisfactory, with the classification accuracy of 86.5%, sensitivity 80%, specificity 91.57% and AUC 0.858. It is also found that tremor, dressing difficulty, speech difficulty, RBD questionnaire score, and age are important in differentiating early stage PD from healthy control.
|Journal||Journal of Physics: Conference Series|
|Publication status||Published - 7 Jan 2021|
|Event||10th International Conference and Workshop on High Dimensional Data Analysis, ICW-HDDA 2020 - Sanur-Bali, Indonesia|
Duration: 12 Oct 2020 → 15 Oct 2020