Abstract
The University of Central Florida invention introduces a cost-effective smart shoe system for estimating human kinematics and kinetics to produce a physics-informed deep-learning model. Developed to provide accurate analysis of human movement, the system uses affordable, embedded sensors, including an inertial measurement unit and one to detect shoe sole pressure.
Traditional deep-learning models struggle with inconsistent and heterogeneous movements, especially in individuals with neurological disorders. The UCF approach combines interpretable dynamic equations of motion with deep learning algorithms, enhancing reliability and accuracy. The smart shoe has many applications, such as clinical diagnostics for pathological walking, safety evaluations at construction sites, athletic performance monitoring, and general health assessments to identify falls and disease progression.
The significance of gait analysis extends beyond merely identifying abnormalities; it has evolved into a multidisciplinary research field with implications for improving quality of life and informing clinical decision-making. By scrutinizing gait patterns, researchers and healthcare professionals gain valuable insights into the severity, progression, and diagnosis of various diseases and injuries.
Technical Details: The UCF technology includes a smart shoe apparatus and a method for estimating human kinematics and kinetics. In an example setup, the apparatus has at least one inertial measurement unit (IMU) sensor and a shoe sole pressure sensor. The IMU sensor measures three-dimensional acceleration and angular velocity, while the shoe sole sensor measures pressure distribution across the sole. Based on data from these sensors, a processor executes a physics-informed deep learning model that combines dynamic equations of human motion and deep learning algorithms to estimate joint angles, joint moments (such as knee abduction and adduction), and ground reaction forces (GRFs). The apparatus also includes a communication module to wirelessly transmit the kinematic and kinetic parameters to an external device for further analysis, display, and storage.
Partnering Opportunity: The research team is seeking partners for licensing, research collaboration, or both.
Benefit
Quick setup, fast processing, and reduced sensor requirementsMeasures and analyzes human motion in all conditions encountered in daily lifeEnables measuring progressive neurological diseases (stroke, multiple sclerosis, Parkinson’s disease, etc.)Protects privacy while measuring and analyzing human motionAddresses the limits of current motion capture technologies in terms of time and spaceProvides a cost-effective, efficient solution for comprehensive human motion analysisMarket Application
Shoe companiesHealth data management firmsSports performance analyses institutionsHuman wearable assistive device companiesSoftware companies
Brochure
