SHB: Type I (EXP): Personalized Signal Processing for Early Diagnosis of Mobility Impairment

导致行动障碍的疾病、障碍和退化可能对功能独立性和生活质量产生重大影响。对于许多此类疾病，早期诊断是成功干预和最大限度健康的关键，准确的活动评估是监测进展和评估物理和药物治疗影响的核心。 近年来，惯性体传感器网络（BSN）已经成为一种平台，具有在任何位置提供持续、非侵入性的高精度运动数据收集的潜力。 到目前为止，这些方法主要用于临床，使得从收集的数据中进行绝对评估具有挑战性。门诊外采集会引入讨厌的变量，这会加剧问题。然而，如上所述的移动性评估的目的并不真正取决于与统计标准相比的绝对评估，而是取决于与个体基线相比的相对评估，这消除了干扰变量并使使用BSN的门诊外评估成为更易处理的问题。 该项目利用更个性化医学的一般动机，使用BSN对两种具有移动障碍症状的示例条件进行连续的纵向步态评估：正常压力脑积水（NPH）和多发性硬化症（MS）。具体的预期贡献包括：1）识别可以从临床外惯性BSN数据中可行地提取并有效地用于检测个体化步态变化的信号特征，2）开发用于步态变化的个体化识别和相对量化的新的信号处理算法，3）探索在资源上实现个性化信号处理形式的技术-受限的BSN平台，以实现更智能的数据减少和动态能量优化策略，从而延长此类系统的电池寿命。

Diseases, disorders, and degenerations that lead to mobility impairment can have a significant impact on functional independence and quality of life. For many such conditions, early diagnosis is the key for successful intervention and maximized wellness, and accurate mobility assessment is central to monitoring progression and evaluating the impact of physical and pharmaceutical therapies. In recent years, inertial body sensor networks (BSNs) have emerged as platforms with the potential to provide continuous, non-invasive collection of high-precision motion data in any location over an extended period of time. Up to now these have been used primarily in-clinic, making absolute assessments from data collected challenging. Out-of-clinic collections introduce nuisance variables, which exacerbate the problem. However, the purpose of mobility assessments as described above does not truly depend on absolute assessments compared to a statistic norm but rather on relative assessments compared to an individual's baseline, which eliminates nuisance variables and makes out-of-clinic assessments using a BSN a more tractable problem. This project leverages the general motivation of more personalized medicine to enable continuous, longitudinal gait assessments to be made using a BSN for two example conditions with mobility impairment symptoms: normal pressure hydrocephalus (NPH) and multiple sclerosis (MS). Specific anticipated contributions include: 1) identifying signal features that can be feasibly extracted from out-of-clinic inertial BSN data and effectively utilized for detecting individualized gait changes, 2) developing new signal processing algorithms for the individualized identification and relative quantification of gait changes, and 3) exploring techniques for implementing forms of personalized signal processing on resource-constrained BSN platforms to enable more intelligent data reduction and dynamic energy optimization strategies that can extend the battery life of such systems.