UC Berkeley

Prosthetic limbs and assistive devices aim at returning functionality and improving quality of life for people with limb differences. They generally require customization to effectively meet the specific needs of their users. Despite the expense and effort involved in acquiring such a prosthetic, many people with limb loss end up abandoning their devices. Acceptance is contingent upon the comfort of the user, aesthetics, and functionality of the device.

The overall goal of my doctoral dissertation was to advance the research and development of low-cost upper-limb assistive devices. Towards this goal, my interest was also sparked in the examination of new concepts of variable stiffness technologies relevant to the mechanical design and function of such prosthetics. The guiding premise was to provide customized and affordable assistive devices to enhance the ability of people with upper-limb differences not only to perform activities of daily living, but also to help them enhance their quality of life. In other words, the idea was to `adjust the device to the user rather than the user to the device'. Thus, by its very nature, this problem is complex and multi-faceted and as part of my work three specific research and development directions have been advanced: the mechanical design and evaluation of (1) a parametrically designed prosthetic finger, (2) an assistive exoskeleton glove, and (3) a semi passive prosthetic for playing the drums.

In designing a 3D printed prosthetic finger, an important consideration was the comfort of the user and the ability to adjust the device to their specific needs. Recognizing that most current 3D printed prosthetic designs available online use uniform scaling to fit prostheses, compromising adjustability, we proposed a parametric modeling method for design. We showed that a prosthetic finger designed using parametric modeling has a range of motion (ROM) (path of the fingertip) that closely aligns with the finger's would-be natural path. We also showed that the ROM produced by a uniformly scaled prosthetic poorly matches the ROM of the finger. To further argue for the need of parametric designs, finger width and length measurements were collected from 50 adults between the ages of 18-30 and was found that there is negligible correlation between length and width of the index finger. From the highest and the lowest length to width ratios of index fingers found among participants, we built a prosthetic finger using two different design methods: uniform scaling and parametric modeling. The mechanical designs utilized a ``crossed four bar'' linkage mechanism, and its ROM was determined by Freudenstein's equations. By simulating the different paths of the fingers, we demonstrated that parametrically modeled fingers outperform uniformly scaled fingers at matching a natural finger's path.

An exoskeleton glove was designed to assist people with limited finger mobility and/or grip strength, which can result from accident, neurodegenerative diseases or even deteriorated hand functionality due to aging. The intended function of the orthosis was to support and enable light intensity activities of daily living (ADLs) and improve the ability to grasp and hold objects. The goal was also do develop a low cost, open source, 3D printed device, that can be easily manufactured and assembled. The proposed hand exoskeleton, utilizes a cable-driven design applied to individual digits with motors, and incorporated the Sparthan module which enables to electronically control each finger independently, with incorporated current detection for feedback control. Two different methods were suggested for motor actuation: the Myo Armband and flex sensors. Evaluation of durability and performance efficacy was performed in a case study, pointing out some shortcomings but also promising paths for future improvements.

The development of a variable stiffness prosthetic for playing the drums occupied a significant part of my work. The goal was to develop a prosthetic device to allow a drummer with unilateral wrist articulation or transradial amputation to play using the remaining functioning arm and the prosthesis. Along these lines, the objective was to replicate finger and wrist function for drumming, and examine whether an adjustable prosthetic device can improve user dexterity and coordination, and reduce muscle fatigue. The hypothesis was posed that variable stiffness in a prosthetic for drumming will increase drumming accuracy of double strokes at a range of tempi, and reduce muscular fatigue and perceived effort. This hypothesis was examined through a user study involving 9 participants with and without upper-limb differences, empirically through discussions with users about feel and perception of the device, and objectively through a user study in which signals were collected and analyzed for comparison and assessment. The study involved a rhythmic synchronization test of a common drum pattern at varying tempi, where several metrics of measured performance and muscle activity using surface electromyography (sEMG) were chosen to evaluate accuracy and fatigue. Despite the complexity of the problem, some advantages of variable stiffness settings compared to a rigid setting were confirmed and quantitative insights were gained on the trade-offs between accuracy and muscular effort. The full extent of utilizing variable stiffness control remains to be further evaluated by long-term studies, which will test the adaptability of the user to learn and utilize the energy storing capabilities of a variable elastic element in a prosthetic.

Overall, my research combined my passion for engineering, my love of music, and my desire to use my skills to make a difference in the lives of people around me, to conceive, design and evaluate affordable and customizable prosthetics. It gave me an opportunity to work with engineering fundamentals and explore how engineering principles can be combined in creative ways to tackle complex and multifaceted problems. This work also enhanced my initiative to carry out ideas from paper, to the lab, to a final product, and presented me with opportunities of working with doctors, patients and prosthetic users on a personal level. Although much work remains to be done in the vast field of prosthetics and orthotics, this research has added a valuable contribution to the emerging field of affordable, 3D printed, user-centered devices.