Mapping ankle-foot stiffness to socket comfort and pressure using a robotic emulator platform to personalize prosthesis function via human-in-the-loop optimization

这项初步研究的主要目的是描述假肢踝足僵硬之间的关系， 用户报告退伍军人经胫骨截肢后的舒适度和残端-插座接口压力，并使用 这些关系指导假体优化，以最大限度地提高日常舒适性。用户舒适度是最重要的 对假肢使用者的重要性，并直接影响对假体的满意度、使用或拒绝 假体，以及最终的机动性和独立性。在美国，近90%的假体临床应用 与下肢相关，占所提供假肢介入的绝大多数。然而， 调查显示，33%到57%的假肢使用者表示对舒适性不满意，而 佩戴假肢，39%的人表示舒适和合身是他们对新假肢最关心的问题 这也是更换假肢医生的主要原因。此外，51%和37%的越南退伍军人和OIF/OEF 服役人员报告了与假体相关的疼痛。重要的是，定期使用假肢和对 它的舒适性可以增加失去肢体后重返工作岗位的可能性，这是一个关键因素 退伍军人重新融入社区。自我感知的舒适度是一个多因素的、动态的、心理物理的结构， 但有证据表明，它受到残渣-插座界面压力的影响。这些界面压力是 受承窝设计的影响，也受假体对位给出其对地面反作用力传递的影响 强行穿过眼窝至近端解剖。通过相同的机制，假体的刚度也应该 理论上影响界面压力，但这种关系尚未量化。为此，假肢 可以对属性进行调整，以操纵界面压力以实现舒适性最大化。因此，具体目标是 本研究的主要内容是：1)定义假肢足刚度、残端-承窝界面压力、 和用户感知的舒适性，以及2)评估人在环中优化以调整假肢的效果 将界面压力降至最低的硬度独立于临床优化的插座。 我们将通过使用新型机器人假肢仿真器平台来实现研究目标，该平台包括一个高性能的 高性能机电一体化系统，用于快速、可控和高分辨率的龙骨刚度调制 系留假肢踝足末端效应器。这两个目标都将涉及10名患有单侧经胫骨的参与者 截肢。对于目标1，参与者将首先对仿真器系统进行装配和适应，并 协议，然后在不同的假体脚踝-足部僵硬设置下以稳定状态行走 特定于他们的身体质量的参考值的某一百分比，该百分比反映了共同的动态弹性 回应脚步。峰值界面压力和插座舒适性将在每个刚度设置下进行测量，以量化 通过线性混合模型和曲线评估的刚性、舒适性和压力之间的关系 很合身。对于目标2，参与者还将经历对仿真器系统的装配和适应，然后 体验一个称为人在环优化的过程，在这个过程中，假体的龙骨刚度将 使用侧重于最小化界面的贝叶斯控制算法实时自动优化 压力成本函数和最大化舒适性。除了对参与者体验的持续反馈之外， 峰值压力、插座舒适性和感觉到的作用力将在优化的刚度设置之间进行比较 和目标1的参考刚度设置。[对于这两个目标，将测量下肢运动学以 评估僵硬对步态表现的影响。]这项研究的结果将为临床上重要的 假体硬度、界面压力和舒适性之间的关系以指导处方指南 最大限度地提高经验丰富的假肢使用者的行走舒适性。我们的成果也将为发展奠定基础 通过未来的优秀奖自动实现僵硬调整的智能假体 对生物反馈施加压力，以维持退伍军人长期的日常舒适性、假体使用和独立性。

The primary aim of this pilot study is to characterize the relationships between prosthetic ankle-foot stiffness, user reported comfort, and residuum-socket interface pressure in Veterans with transtibial amputation, and use these relationships to guide prosthesis optimization to maximize daily comfort. User comfort is of paramount importance to leg prosthesis users and has a direct impact on satisfaction with a prostheses, use or rejection of a prosthesis, and ultimately mobility and independence. Nearly 90% of prosthetic clinical encounters in the US are related to the lower limb, representing the vast majority of delivered prosthetic device interventions. However, surveys suggest that between 33% and 57% of leg prosthesis users report dissatisfaction with comfort while wearing their prosthesis, and 39% indicated that comfort and fit were their biggest concerns with a new prosthesis and a primary reason for changing prosthetists. Moreover, 51% and 37% of Vietnam Veterans and OIF/OEF Service Members reported prosthesis related pain. Importantly, regular use of a prosthesis and satisfaction with its comfort can increase the likelihood of returning to work following lower limb loss and this is a critical factor to Veterans’ community reintegration. Self-perceived comfort is a multifactorial, dynamic, psychophysical construct, but evidence suggests it is influenced by residuum-socket interface pressures. These interface pressures are affected by socket design, but also by prosthesis alignment given its influence on transfer of ground reaction forces through the socket to proximal anatomy. By the same mechanism, prosthesis stiffness should also theoretically affect interface pressures, but this relationship has not been quantified. To this end, prosthesis properties could be tuned to manipulate interface pressures for maximizing comfort. Therefore, the Specific Aims of this study are: 1) Define the maps connecting prosthetic foot stiffness, residuum-socket interface pressures, and user-perceived comfort, and 2) Assess the efficacy of human-in-the-loop optimization to tune prosthetic foot stiffness for minimizing interface pressure independent and in addition to the clinically optimized socket. We will address the study aims through use of novel robotic prosthesis emulator platform that includes a high- performance mechatronic system for rapid, controlled, and high-resolution keel stiffness modulations of a tethered prosthetic ankle-foot end effector. Both aims will involve ten participants with unilateral transtibial amputation. For Aim 1, participants will first undergo fitting and accommodation to the emulator system and protocol, and then walk at steady state under different prosthesis ankle-foot stiffness settings stratified by a certain percentage from a reference value specific to their body mass that reflects a common dynamic elastic response foot. Peak interface pressures and socket comfort will be measured at each stiffness setting to quantify associations between stiffness, comfort, and pressures as assessed through linear mixed modelling and curve fitting. For Aim 2, participants will also undergo fitting and accommodation to the emulator system, and then experience a process known as human-in-the-loop optimization in which the prosthesis keel stiffness will be automatically optimized in real-time using Bayesian control algorithms focused on minimizing an interface pressure cost function and maximizing comfort. In addition to continuous feedback on participant experience, peak pressures, socket comfort, and perceived effort will be compared between the optimized stiffness setting and the reference stiffness setting of Aim 1. [For both aims, lower extremity kinematics will be measured to assess stiffness effects on gait performance.] Results from this study will inform on the clinically important relationships between prosthesis stiffness, interface pressures, and comfort to guide prescription guidelines for maximizing walking comfort in Veteran prosthesis users. Our results will also set the foundation for development of a smart prosthesis through future Merit Awards that automatically implements stiffness adjustments according to pressure biofeedback to maintain long-term Veteran daily comfort, prosthesis use, and independence.