One of the many services that intelligent systems can provide is the ability to analyze the impact of different medical conditions on daily behavior. This study uses smart homes and wearable sensors to collect data while (n=84) older adults perform complex activities of daily living. The collected data is analyzed using machine learning techniques. Analysis reveals that differences between healthy older adults and adults with Parkinson’s disease not only exist in their activity patterns, but that these differences can be automatically recognized. The machine learning classifiers reach an accuracy of 0.97 with an AUC value of 0.97 in distinguishing these groups. Permutation-based testing confirms that the sensor-based differences between these groups are statistically significant, which offers insights for automatic detection of the behavior impacts of Parkinson’s disease.
The post Analyzing Activity Behavior and Movement in a Naturalistic Environment Using Smart Home Techniques appeared first on Journal of Biomedical and Health Informatics.