View the videos below for a brief introduction to some of our work.
Machine learning work
Presentation of machine learning work at virtual 2020 IEEE International Conference on Bioinformatics and Biomedicine (IEEE BIBM) Machine Learning and Artificial Intelligence in Bioinformatics and Medical Informatics Workshop.
Use of deep learning for automatic classification of independent components from scalp cortical activity recordings.
Use of machine learning to identify community dwelling older adults with balance dysfunction using short duration acceleration data.
Fall Risk Detection Augmented Reality:
Given the high prevalence for falls in community dwelling older adults, the following fall risk detection tool uses Swift, CoreML, and ARKit to identify common objects and provide a rating of fall risk depending on context. This project seeks to build upon ML/AI approaches to improve object recognition at home, and integrate with evidence-based guidelines for reducing fall risk.
Auto Industry Hazards Simulator:
Given the high incidence of musculoskeletal injuries in the workplace, it is crucial to provide a training platform to allow for proper practice of challenging working conditions and manual labor tasks in a safe and well monitored environment, so as to provide feedback on the onset of unsafe strategies that may more readily lead to injury. The project seeks to integrate an HTC Vive together with physical objects used for manipulation and kinetic and kinematic sensors to provide an overview of performance in demanding auto industry tasks.
Virtual obstacle navigation:
Given the high risk of falls in older adults, this project aimed at developing a training platform for avoiding obstacles and multitasking though either eye movement, hand movement, or body movement. It provides a proof-of-concept for different interactions that can be carried out in a virtual environment, aimed around the theme of obstacle navigation. The project seeks to integrate an HTC Vive with kinetic and kinematic sensors to monitor performance in attention-demanding virtual environments.
Virtual paddleboarding v.1:
The goal of this project is to develop an immersive paddleboarding game that motivates those going through physical rehabilitation by testing their balance in a fun way. The project integrates the HTC Vive together with Unity software, motion capture, and a moving platform.
Visual cliffs, virtual reality, and movement disorders:
Given the high risk of falls and prevalence of anxiety in older adults, and particular in those with movement disorders, this project aims to monitor concurrent heart and brain activity while a person is walking on a treadmill and experiences visual cliffs in immersive virtual reality environments. It provides a novel platform for examining real time anxiety changes in older adults and modulate anxiety levels through adaptive exposure. This project integrates the HTC Vive with Unity software, EEG, EKG, and an instrumented treadmill.
Don’t look down:
Given the prevalence of fear of falling in older adults, this project aims to provide an environment that allows for the exploration of changes in movement due to walking on an elevated platform. This project provided a proof-of-concept for the integration of treadmill data within a virtual reality environment.