For prosthesis users, knee units can range from simple devices costing $2000 up to $45,000 for high-end, microprocessor-controlled systems. These higher-end electronic knees provide significant advantages in stability, gait, and metabolic rate compared to their passive or mechanical counterparts. However, the high cost of such systems makes them inaccessible to most amputees. In this study, it was hypothesized that a microprocessor knee could be manufactured for less than $1000, with comparable stability and user experience to a high-end industry standard device. A prototype (E-Knee) was designed with a specific emphasis on stance stability, and was tested during patient gait trials. The gait trials used a repeated measures design to compare three knee devices (a simple passive knee, the prototype E-Knee, and a high-end knee). Ground reaction forces and a functionality questionnaire were used to compare devices. A microprocessor locking test was used to evaluate the prototype’s ability to prevent falls. Building on the LIMBS M3, a passive four-bar polycentric device, the E-Knee added sensing, computing, and controlling capabilities for a material cost of $507. Initial data from a two-subject trial served as proof-of-concept to validate the prototype and found that it improved gait by providing more stability than the M3 and had more gait-pattern similarities to the Ottobock C-Leg than to the M3. Patients reported no perceived differences in stability between the E-Knee and the C-Leg. Patient trials supported that the E-Knee prototype behaved more naturally than the low-end M3 device and had similar ground reaction forces to the C-Leg.

Approximately 82% of amputees prefer microprocessor knees (MPKs) to the passive alternatives. However, the cost of these devices makes them inaccessible for many patients. The aim of this research is to develop an affordable MPK that allows for stumble reduction and flexion dampening at a fraction of the cost of similar devices. The GKnee was developed by a sophisticated mathematical model that can effectively calculate geometric configuration and simulate forces transferred through a prosthetic knee at any given point through the gait cycle. With a median error of 6%, the mathematical model was developed to the point of reasonable accuracy for determining component placement and force interactions. The model served as a valuable tool to assist in the iterative design process of the GKnee, influencing component selection for the hydraulic system and frame design. This model was then validated using a compression rig and a mock GKnee prototype. The GKnee was then evaluated for its ability to perform under expected loading conditions, using compression testing and dynamic flexion testing. This research led to the development of a sub USD 500 microprocessor prosthetic, while remaining under 2.27 kg.

Objective: To train a machine-learning (ML) algorithm to classify stumbling in transfemoral amputee gait. Methods: Three subjects completed gait trials in which they were induced to stumble via three different means. Several iterations of ML algorithms were developed to ultimately classify whether individual steps were stumbles or normal gait using leave-one-out methodology. Data cleaning and hyperparameter tuning were applied. Results: One hundred thirty individual stumbles were marked and collected during the trials. Single-layer networks including Long-Short Term Memory (LSTM), Simple Recurrent Neural Network (SimpleRNN), and Gradient Recurrent Unit (GRU) were evaluated at 76% accuracy (LSTM and GRU). A four-layer LSTM achieved an 88.7% classic accuracy, with 66.9% step-specific accuracy. Conclusion: This initial trial demonstrated the ML capabilities of the gathered dataset. Though further data collection and exploration would likely improve results, the initial findings demonstrate that three forms of induced stumble can be learned with some accuracy. Significance: Other datasets and studies, such as that of Chereshnev et al. with HuGaDB, demonstrate the cataloging of human gait activities and classifying them for activity prediction. This study suggests that the integration of stumble data with such datasets would allow a knee prosthesis to detect stumbles and adapt to gait activities with some accuracy without depending on state-based recognition.

The BSc in Engineering Innovation and Leadership (E-Lead) at The University of Texas at El Paso (UTEP) is an ABET-accredited engineering degree that is transforming how engineering is learned and taught. At its core, E-Lead is an integrated engineering degree that provides students with a diverse foundation in multidisciplinary engineering knowledge, leadership theory and practice, innovation skills, and an understanding of business. The degree is customisable, allowing students to couple their E-Lead education modules with a focus on a specific traditional engineering discipline. This content knowledge is then integrated and applied through the E-Lead core modules each semester, enhanced by peer and faculty support and extracurricular programming offered by the department. As an undergraduate degree, team-based, project-based learning is the cornerstone of how we support our students in developing transferable competencies and skills. This project work is facilitated in our dedicated studio space, where students not only attend class but can collaborate on campus after hours and use the available prototyping lab to work on their projects. Therefore, the E-Lead degree focuses on developing a learning environment where transferable skills flourish, using real-life scenarios and projects to prepare students to take the initiative to ‘engineer’ the future and solve socio-technical challenges. The chapter concludes with practice-focused recommendations for designing integrated engineering leadership curricula, including strategies for scaffolding transferable skills, aligning leadership development with technical coursework, and engaging industry partners.