Purple Mountain’s Majesty Team 2020
Fall Prediction using Anomaly Detection in Gait Patterns
A fall can be one of the most detrimental events an elderly individual can experience, often resulting in an exponential deterioration in health. Such repercussions include injuries like a hip fracture, traumatic brain injury, upper limb injury; post fall syndrome effects like immobilization, depression, confusion, and loss of autonomy and dependence. Fall alert systems are already on the market, but our solution allows for fall prediction, preventing not only a jurassic event from occurring, but also the repercussions to quality of life that follow. If we monitor a user’s walking patterns and understand when their trends are reminiscent and indicative of a fall, we can notify them before the event happens. We created a smart insole that relies on measurements suggested by podiatrists and occupational therapists to watch for abnormal gait patterns and prevent falls before they happen.
Related Tech Notes
- Anomaly detection using unsupervised learning by Mateo Guaman
- Fall Prediction using Anomaly Detection by Peter Malinovsky
- Capacitive Sensors: How do they work and what can they do? by Kevin Naranjo
- Capacitance Based Pressure Sensors by Ashwin Swar
Project Poster
