Method and system for analyzing correlations between motor behavior and cognitive ability

Abstract

The invention provides a method for analyzing correlations between motor behavior and cognitive ability. The method includes the steps of extracting a motion feature set from motor behavior data, wherein the motion feature set reflects behavior features; adopting an iterative random forest to select an optimal feature subset for detecting the cognitive ability from the motion feature set; adopting the optimal feature subset to acquire analysis results of the correlations between the motor behavior and the cognitive ability by the adoption of a classifier. According to the method for analyzing the correlations between the motor behavior and the cognitive ability, the correlations between the motor behavior and the cognitive ability can be effectively analyzed, and especially for a data set with unbalanced small samples, the recognition rate of patients suffering from cognitive disorders can be increased.

Description

technical field [0001] The invention relates to the field of machine learning, in particular to a method and a system for analyzing the correlation between motor behavior and cognitive ability. Background technique [0002] Neurodegenerative disease is one of the major chronic diseases that threaten the life and health of the elderly. The traditional medical point of view divides neurodegenerative diseases into two categories: one is cognitive impairment diseases, and patients with such diseases may have impairments in memory, language or memory, while their motor functions are preserved; the other is Movement disorders are disorders in which motor function is impaired but cognitive function is fully preserved. In recent years, many clinical medical studies have shown that cognitive impairment is usually accompanied by motor dysfunction, and motor dysfunction is often accompanied by cognitive dysfunction. [0003] In the prior art, in the assessment of cognitive ability, t...

AI Technical Summary

Problems solved by technology

[0005] However, these methods in the prior art still have some deficiencies in generality, pertinence and applicability

For example, in the analysis of the association between motor behavior and cognitive ability, it is usually difficult to collect data on a large scale, and because the number of patients with cognitive impairment is usually much smaller than that of patients without cognitive impairment, the collected data is also characterized by imbalance , traditional machine learning algorithms are usually not suitable for small sample data sets with unbalanced characteristics, therefore, leading to problems such as low recognition accuracy of small categories (patients with cognitive impairment)

Images