Intelligent video surveillance to detect mobility problems of older adults at home: from fall detection to fall prevention

点击翻译按钮获取中文摘要

According to the World Health Organization, 30% of the population will be seniors (65 years and over) in 2050. This major demographic transformation has important repercussions since many events can affect the health, security and quality of life of these more vulnerable people. For instance, the greatest hazard for older adults is accidental falls, and 85% of injury-related hospitalizations for seniors are the result of falls. Innovative approaches need to be developed and implemented to detect, prevent and mitigate these impacts. Our general long-term objective is to investigate, develop and evaluate low-cost intelligent video surveillance systems (IVS) to support older adults at home or in senior housings for improving their quality of life. This will be pursued during this 5-year grant by focusing on the prevention of accidental falls. In addition to fall detection, we will develop artificial intelligence algorithms (e.g. deep learning) to detect early signs of mobility decline in older adults such as near-fall events (e.g. stumble, slip, misstep) or abnormal walking (e.g. unsteadiness, unusual gait speed) since they are considered as good predictors of future falls. Continuous monitoring of fall risk with the IVS will help to quickly identify mobility problems for more efficient intervention before a fall occurs with its negative consequences (e.g. fractures and other injuries, fear of falling). This system will also offer an economic advantage by supporting the older adult at home which is much less expensive than institutionalization. The IVS will advantageously complement other assisted living devices such as ambient and wearable sensors for the benefit of seniors.