Menu Close

Anti-Fall Strategies

Pre-Impact Fall Detection – ASCane & ASBGo SW

 

The high number of fall-related injuries are one of the most common health concerns especially in elderly people, being a marker of frailty, acute and chronic health impairment. According to the World Health Organization (2018), falls are the second main reason of death by accident worldwide, killing an estimated 646 000 individuals per year. Moreover, 37.3 million falls are severe enough to require medical attention each year. As natural effects after a fall comes the fear of falling again, depression, social isolation, physical decline, or feelings of helplessness. Slow interventions after a fall may exacerbate the whole situation and the mentioned effects. Also, the treatment of fall-related injuries has economic repercussions. Only in the United States of America, in 2000, $19 billion were spent on the direct medical costs of fall-related injuries. In 2015, this value has risen to more than $31 billion in the Medicare alone, and by 2020, it is expected to surpass the $40 billion barrier.

These reasons lead us to believe that it is important to invest in solving this problem. However, the majority of falls occur due to more than one single cause. Multiple interactions between a subject with great risk of falling and mediating factors are responsible for falls or alterations of balance. For example, neurological and muscular diseases such as Parkinson’s disease (PD), Dementia (Dm), Stroke, or Osteoarthritis can increase the risk of falling. Medications such as antidepressants, sedative hypnotics or neuroleptics are potential reversible risk factors for falls in the elderly. Even hypoglycemic agents, visual impairment, postural hypotension and vitamin D insufficiency have been implicated as risk factors for falls in a few retrospective studies, as well as footwear or environment hazards, including wet floors, poor lightning or improper bed height.

This Ph.D. project focuses mainly on the following cause: the non-use and non-access to fall-related monitoring devices. In one hand, the majority of these devices are mainly based on real-time fall detection, which means they cannot avoid a fall, and they are usually expensive. On the other hand, they had not been so disseminated among the community and they are usually rejected because they are difficult to use in Ambient Assisted Living (AAL), e.g. monitoring devices based on wearable sensors.

Thus, the main goal is to contribute with the development of anti-fall strategies and safety measures that can work in collaboration with daily life accessories mainly used among the elderly and patients with gait disorders. It is intended to monitor subjects continuously, detect abnormal situations with a high risk of falling, and act to prevent falls by avoiding them or at least minimize related injuries. At the moment, focus has been on canes and walkers due to their popularity and usability within the mentioned population.

Assistive Smart Cane (ASCane)

The ASCane concept emerges as a solution to the presented problem, and describes a unique and advanced cane capable of monitoring its user in real-time with fall-related strategies. Its main characteristic is the actuator system that secures stability to the user when the user has imbalance moments or near-falls. As a first step, suitable physical quantities are measured using appropriate sensors: an Inertial Measurement Unit (IMU); a Force Sensor Resistor (FSR); an ultrassound sensor; and a photodetector. Subsequently, data acquired is analysed, relevant features are computed, and decisions are made by using Artificial Intelligence (AI). Finally, the outcomes of the implemented algorithms will be used in the activation of several mechanisms: Anti-Fall Mechanism, Haptic Feedback Mechanism and Light Sensing Mechanism.

 

ASBGo Smart Walker

The main goal is to define and implement a new and more effective strategy, i.e., an incremental service and innovation to prevent a fall event and improve the quality of rehabilitation in the ASBGo Smart Walker. From all implemented strategies in several smart walkers the final outcome is always stopping the walker when a possible fall is detected. At the same time, the presented systems have only one mode of use, i.e., a defined posture is required. The idea is to implement different strategies for different modes of use by using a set of sensors to monitor the user. Similarly to the ASCane, AI will be used to detect abnormal situations with a high risk of falling and act to prevent falls or near-falls. The visual feedback will help and correct the user.