Longitudinal home-based monitoring of bio signals of bulbar dysfunction to aid prognostication and therapeutic monitoring in MND. The home-based devic

我们的目标是开发一种非侵入性生物信号记录设备，设计用于MND患者在家中收集纵向数据，并创建一个人的球功能的数字双胞胎模型，重点是吞咽。我将根据研究者Johnny McNulty在之前的研究中使用的BIOPAC金标准设置验证肌电肌传感器和麦克风。我还将针对加速计MMR-MOTION R验证IMU，该加速计也在早期研究中使用。验证工作将在健康志愿者中进行。一旦通过验证，我将组织一个焦点小组来讨论设备的设计和预期用途。在这个共同设计阶段之后，我将进一步开发电子和电源管理，使设备便携和可穿戴。我还会制作一些材料来教病人和护理人员如何使用它。注：这不是医疗器械，而是用于科学研究的器械。它的使用需要伦理批准，但不需要MHRA批准。与临床MND专家James Bashford和患者代表合作，我们将创建一个任务列表，我们希望用户在佩戴传感器时执行这些任务。这些将是简单的任务，容易和快速进行定期在家里收集纵向数据。首先，我将使用原始数据来计算平均值、标准差和峰值，观察它们如何随时间演变。然后，我将利用这些信息开发一个机器学习模型，将生物信号数据与症状联系起来。然后，我将用它来创建一个预测系统，它可以识别参数的变化，预测症状的演变。在每周从患者家中收集数据的同时，患者还将被要求每月一次到医院或实验室进行高密度表面肌电图、声门电图和肺活量测定仪的进一步检查。这些方法已经在发表的论文中使用过，我还需要确定我将使用哪种设备。

We aim to develop a non-invasive bio-signal recording device designed to be used at home by people living with MND to collect longitudinal data and create a digital twin model of a person's bulbar function focused on swallowing.I will validate the EMG muscle sensors and microphone against a gold standard setup by BIOPAC that researcher Johnny McNulty used in a previous study. I will also validate the IMU against an accelerometer MMR-MOTION R, also used in an earlier study. The validation work will be performed with healthy volunteers.Once validated, I will run a focus group to discuss the device's design and intended use. Following this co-design stage, I will further develop the electronics and power management to make the device portable and wearable. I will also create material to teach patients and carers how to use it. Note: this is not a medical device, but a device used for scientific research. Ethics approval will be required for its use but not MHRA approval.Working with clinical MND expert James Bashford and patient representatives, we will create a list of tasks that we intend the users to carry out while wearing the sensors. These will be simple tasks, easy and quick to carry out regularly at home to collect longitudinal data. First, I will use the raw data to calculate mean, standard deviation, and peak values to observe how they evolve with time. Then, I will use that information to develop a machine-learning model that relates bio-signal data with symptoms. I will then use this to produce a prediction system that identifies parameter changes predictive of symptom evolution.While collecting data weekly at home from the patients themselves, the patients will also be asked to come once a month to the hospital or labs for further tests using high-density surface EMG, Electroglottograph and spirometry. These methods have been used in published papers and I still need to finalise which equipment I will be using.