Objective: To differentiate patients with Parkinson’s disease (PD) with and without mild cognitive impairment (MCI) using quantitative gait variables through a data mining approach

Background: Cognition and gait in Parkinson’s disease (PD) appear to be closely related in a complex fashion. Previous studies (1) have shown that dysfunctions in specific gait parameters are associated with cognitive decline in PD. Originally conceptualized as the transitional state between normal aging and Alzheimer’s disease, the construct of MCI has been recently used in PD, as PD-MCI, to identify a picture of cognitive decline without impaired functional activity (2)

Method: We consecutively evaluated cognition and gait in 27 patients with PD. Cognitive performance was evaluated with a neuropsychological battery assessing memory, executive/attention, and visuospatial domains. Gait was investigated using a gait analysis system in the following conditions: 1) normal gait; 2) motor dual task; and 3) cognitive dual task. Smote technique (3) and Random Forests were implemented in order to make the number of records rise and to predict the presence of MCI

Results: Twenty-two gait features were included in the analysis after a selection through the computation of a matrix and the choice of a threshold as regards correlation. After splitting dataset into training and test sets, evaluation metrics were computed. Accuracy, specificity, sensitivity, recall and precision in predicting MCI resulted all over 90%

Conclusion: The present pilot study shows that a data mining approach using gait analysis variables exhibits high accuracy, specificity and sensitivity in predicting the presence of MCI in PD, thus suggesting specific connections between gait and cognition in PD. In addition, our results prompt that a data mining approach on gait parameters may represent a reliable surrogate biomarker of cognitive impairment in PD

References: 1. Amboni M et al. Gait patterns in Parkinsonian patients with or without mild cognitive impairment. Mov Disord. 2012 Oct;27(12):1536-43. 2. LItvan I et al. Diagnostic criteria for mild cognitive impairment in Parkinson’s disease: Movement Disorder Society Task Force guidelines. Mov Disord. 2012 Mar;27(3):349-56 3. Chawla N. et al. (2002). SMOTE: Synthetic Minority Over-sampling Technique. JAIR. 2002;16: 321-357.

« Back to 2019 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/gait-patterns-may-distinguish-parkinsonian-patients-with-and-without-mild-cognitive-impairment-a-data-mining-approach/