Team 27

Cerebelarm: Machine Learning Enabled Vision Assisted Pediatric Prosthetic

Team Members: Danielle Carpenter, Daniel Brennen Martin, Dylan Mitchell, Justin Pettit
& Juan Pablo Robayo

Team Mentors: Marco Santello, PhD - SBHSE

YouTube Link: View the video link below before joining the zoom meeting

Zoom Link: https://asu.zoom.us/j/85420274788

Abstract

Approximately 10,300 children are born with a congenital, upper limb deformity annually in the United States. Additionally, 58% of upper limb prosthetic users do not utilize their prosthetic on a regular basis. Numerous articles of literature state that children are far more adaptable to change, particularly at a younger age. Therefore, children would be far more open to the acceptance of a prosthetic lifestyle, normalizing it in modern society. This is where the need for pediatric prosthetic limbs is taken into account. Although there are many forms of prosthetics on the market, they can be expensive and hard to adjust as a child grows. Not only this, but current models of upper limb prosthetics require the use of a manually cycling grip types to pick up objects limiting the intuition and range for the usage of the device. To combat these challenges, the Cerebelarm team has created a concept of a pediatric, upper limb prosthetic that is made through the use of additive manufacturing. Furthermore, this device utilizes an integrated optical sensor that provides object recognition to dynamically change the grip type based on the object in view. Through this first semester of product design, the team has identified key product specifications to develop a durable, lightweight, water resistant, vision enabled prosthetic that can be scaled with a child’s growth between ages 5 and 18, our targeted market, which can be quickly customized to the user’s aesthetic preferences. This design establishes itself among the competitors as a more intuitive, everyday device that parents and users can more readily adapt to without worrying about damage from impacts or water exposure. The device would be classified as a FDA class 2 device due to the presence of a microcontroller. The plan for the future is to prototype, characterize build materials and processes, and develop the code necessary to functionalize the object recognition and its integration into the overall arm control. The team aims to have a market price between $3,000 and $8,000 with a half price reduction for subsequent arm refits. The Cerebelarm team is confident that the proposed device will give kids an arm they can identify with, allowing them to live their lives with minimal extraneous thought. Our vision is that users of our device will more abundantly integrate prosthetics with their identity so that they might be more open to larger prosthetic innovation in the future.