Development of a method for monitoring the balance of people with Multiple Sclerosis using smartphone inertial sensors

多发性硬化症会对中枢神经系统造成损害，通常会导致行动能力和保持平衡能力的挑战（也称为姿势稳定性的丧失）。通过评估多发性硬化症患者的平衡，临床医生和研究人员可以更好地了解症状以及它们如何随着疾病进展而变化。现有的临床评估量表可能是主观的，并且在诊所环境中只能提供一次症状的快照，并且患者自我报告的特征（如跌倒次数）已知是不可靠的。其他监测平衡和稳定性的技术依赖于运动捕捉摄像机、测力板和压力传感器等基础设施，这些技术需要财政投资，并且受到评估病人的速度的限制。惯性测量传感器系统以前被用于监测患有疾病的个体的运动。该项目旨在开发、改进和评估基于智能手机的多发性硬化症患者平衡测量的有效性。凭借其低成本和广泛的可用性，智能手机惯性测量单元中的传感器可以实现近乎持续的监测，并提供近乎实时的数据潜力。工业合作伙伴，F. Hoffmann-La Roche AG的pRED信息学-数字生物标记组，与普利茅斯大学合作进行了一项试验，探索多发性硬化症患者的症状，以及如何使用智能手机监测他们（www.isrctn.com/ISRCTN15993728），这将是该项目中使用的数据来源。这项工作将首先开发一个数据处理管道，使用信号处理从智能手机惯性传感器数据中提取临床相关的平衡特征。力板以前被用于评估多发性硬化症患者的平衡，这项工作将尝试使用智能手机惯性传感器生成合成力板测量，减少对昂贵基础设施的需求。通过建立基于运动捕捉数据的模型和模拟虚拟惯性测量单元的放置，可以探索和优化传感器的放置。被动监测数据将被分析，以确定是否可以在不需要进行特定测试的情况下，在个人的日常生活中检测到多发性硬化症患者有临床意义的平衡特征。最后，这项工作将研究是否可以使用类似的技术来评估MS患者行走时的动态平衡，借鉴控制工程和机器人技术的想法。通过使用大多数智能手机中已经包含的传感器和仪器，这项工作将使监测与平衡相关的多发性硬化症症状变得更容易，这意味着不需要新的设备。创造一种监测姿势稳定性丧失症状进展的新工具将为患者和临床医生提供有用的数据，以告知患者病情的管理。该工具还可用于研究，在研究访问之间监测试验参与者的症状，以了解正在调查的干预措施的效果。这项工作是在“医疗保健技术”和“工程”EPSRC研究主题，工作是在：“辅助技术，康复和肌肉骨骼生物力学”，“数字信号处理”，“传感器和仪器仪表”和“生物信息学”的研究领域，此外，这项工作将借鉴“非线性系统”，“控制工程”和“机器人技术”的思想。

Multiple Sclerosis causes damage to the central nervous system, often resulting in challenges with mobility and the ability to maintain balance (also described as loss of postural stability). By evaluating the balance of people with Multiple Sclerosis, clinicians and researchers can better understand the symptoms and how they change with disease progression. Existing clinical assessment scales can be subjective and offer only a snapshot of symptoms at one time in a clinic setting and patient self-reporting of features such as number of falls is known to be unreliable. Other techniques for monitoring balance and stability rely on infrastructure such as motion capture cameras, force plates & pressure sensors, which require financial investment and are limited by the rate at which they can assess patients. Systems of inertial measurement sensors have previously been used to monitor the movement of individuals with diseases. This project aims to develop, improve, and assess the validity of smartphone-based measures of balance of people with Multiple Sclerosis. With their low cost and widespread availability, the sensors in a smartphone's inertial measurement unit can allow for near constant monitoring with near real time data potential available. The industrial partners, F. Hoffmann-La Roche AG's pRED Informatics - Digital Biomarkers group, have worked with The University of Plymouth on a trial exploring the symptoms of people with multiple sclerosis and how they can be monitored using smartphones (www.isrctn.com/ISRCTN15993728), which will be the source of the data used in this project. This work will begin with developing a data processing pipeline using signal processing to extract clinically relevant features of balance from smartphone inertial sensor data. Force plates have previously been used to assess balance in people with multiple sclerosis, and this work will attempt to generate synthetic force plate measurements using smartphone inertial sensors, reducing the need for expensive infrastructure. By building models based on motion capture data and simulating virtual inertial measurement unit placements, sensor placement can be explored and optimised. Passive monitoring data will be analysed to see if clinically meaningful features of balance in people with Multiple Sclerosis can be detected in an individual's daily life without needing to perform specific tests. Finally, this work will investigate whether similar techniques can be used to assess dynamic balance in people with MS when they are walking, drawing on ideas from control engineering and robotics. This work will make it easier to monitor the symptoms of Multiple Sclerosis related to balance by using sensors and instrumentation already included in most smartphones, which means no new equipment will be required. The creation of a new tool for monitoring the progression of symptoms of loss of postural stability will provide patients and clinicians with useful data to inform the management of the patient's condition. The tool can also be used in research to monitor the symptoms of trial participants between research visits to understand the effect of interventions that are being investigated. This work is in the "Healthcare technologies" and "Engineering" EPSRC Research themes, the work is in the: "Assistive technology, rehabilitation and musculoskeletal biomechanics", "Digital signal processing", "Sensors and instrumentation", and "Biological informatics" research areas, additionally this work will draw on ideas from "Non-linear systems", "Control engineering", and "Robotics"..