Detection of Alzheimer’s Disease in Elder People Using Gait Analysis and Kinect Camera
Introduction: Gait analysis through using modern technology for detection of Alzheimer's disease has found special attention by researchers over the last decade. In this study, skeletal data recorded with a Kinect camera, were used to analyze gait for the purpose of detecting Alzheimer's d...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | fas |
| Published: |
Kerman University of Medical Sciences
2019-12-01
|
| Series: | مجله انفورماتیک سلامت و زیست پزشکی |
| Subjects: | |
| Online Access: | http://jhbmi.ir/article-1-374-en.html |
| Summary: | Introduction: Gait analysis through using modern technology for detection of Alzheimer's disease has found special attention by researchers over the last decade. In this study, skeletal data recorded with a Kinect camera, were used to analyze gait for the purpose of detecting Alzheimer's disease in elders.
Method: In this applied-developmental experimental study, using a Kinect camera, data were collected for 12 elderly women with Alzheimer's disease and 12 healthy elderly women walking in an oval path. After extracting various features of gait, descriptive analysis was performed to compare the features between the healthy and patient groups. Then, a support vector machine classifier was designed to detect elderly people with Alzheimer's disease.
Results: The comparison of extracted features from skeletal data of gait using Kinect camera in this study indicate that the results are matched with previous findings from systems based on other types of sensors. The accuracy, sensitivity, precision and specificity of system designed in the present study for classifying elders with Alzheimer's disease and healthy elders were 91.25%, 93.4484%, 90.5945% and 93.581% respectively.
Conclusion: In addition to descriptive analysis of gait, by using machine learning methods such as support vector machine classifier, elderly people with Alzheimer's disease can be detected based on features extracted from skeletal data of Elderly people. |
|---|---|
| ISSN: | 2423-3870 2423-3498 |