Millimetre-wave AI-enabled Smart Healthcare monitoring

毫米波AI智能医疗监控

• 批准号: 105007
• 金额: $ 3.88万
• 依托单位: NODENS MEDICAL LTD
• 依托单位国家: 英国
• 项目类别: Feasibility Studies
• 财政年份: 2019
• 资助国家: 英国
• 起止时间: 2019 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=105007
• 关键词: Millimetre; wave; AI; enabled; Smart

"NHS hospitals are experiencing widespread shortages in bed space and front-line staff time, contributing to long waiting times and budget shortages. This is largely because demand for services is outstripping growth. Improved efficiencies and patient flow could release up to £900M per annum in savings, save staff time, and free up facilities. The NHS is tackling this through pilots of real-time location sensing (RTLS) technology, which seeks to track patients, staff and equipment in real-time. This has many potential benefits such as: finding equipment more quickly, identifying available bed space, and monitoring patient-staff contact time. However, currently existing product solutions suffer from a combination of: relatively poor tracking accuracy and reliability, expensive and bulky tags (unpopular with staff and patients), and high infrastructure costs. The project seeks to develop an artificial intelligence (AI)-based millimetrewave sensing technology to provide industry-leading tracking resolution, at low costs and ease-of-use.The millimetrewave (mmWave) regime (approximately 30-300 GHz) encompasses the band of the electromagnetic spectrum between microwaves (Wi-Fi, mobile phones) and infrared. Due to recent technological breakthroughs, mmWave electronics will in coming years be used for automotive radar, next-gen Wi-Fi (WiGig) and mobile communications (5G). The technology also allows us to sense objects with extremely high precision, and detect movements of less than a millimetre. This could include wireless detection of heart beats, as well as fall detection or patient tracking, all done without necessitating any worn tags. While our current prototype sensor is extremely accurate, it does not have the 'intelligence' to process all the data it receives, and can have difficulty distinguishing between different objects; so it may not be able to tell between a patient and a bed, for example.To overcome this shortcoming, NodeNs Medical proposes to use state-of-the-art AI algorithms to give the sensor 'smart' capabilities, allowing it to learn to distinguish between different detected objects. This could be done by recognising the periodic rhythm of a heart beat, or a person's swinging arm or leg.NodeNs Medical will develop a 'smart' AI-enabled mmWave RTLS solution which will:1. track patients, staff and equipment, to identify if staff have visited patients or to track infection spreading,2. monitor for fall detection,3. track vital signs.The device will have market-leading accuracy and 'plug-and-play' capabilities for easy installation, helping the NHS to improve patient outcomes while lowering costs."

NHS医院床位和一线工作人员的时间普遍短缺，导致等待时间过长，预算短缺。这在很大程度上是因为对服务的需求超过了增长。改进的效率和病人流量每年可节省高达9亿GB的成本，节省员工时间，并腾出设施。NHS正在通过实时位置感知(RTLS)技术的试点来解决这个问题，该技术寻求实时跟踪患者、工作人员和设备。这有许多潜在的好处，例如：更快地找到设备，确定可用的床位空间，以及监控患者与工作人员的接触时间。然而，目前现有的产品解决方案存在以下问题：相对较差的跟踪精度和可靠性、昂贵且笨重的标签(不受工作人员和患者欢迎)以及较高的基础设施成本。该项目旨在开发一种基于人工智能(AI)的毫米波传感技术，以低成本和易用的方式提供行业领先的跟踪分辨率。毫米波(毫米波)制度(约30-300 GHz)涵盖微波(Wi-Fi、手机)和红外之间的电磁频谱频段。由于最近的技术突破，毫米波电子产品将在未来几年用于汽车雷达、下一代Wi-Fi(WiGig)和移动通信(5G)。这项技术还使我们能够以极高的精度感知物体，并检测到不到一毫米的移动。这可能包括对心跳的无线检测，以及摔倒检测或患者跟踪，所有这些都不需要任何磨损的标签。虽然我们目前的原型传感器非常准确，但它没有处理它接收到的所有数据的“智能”，而且很难区分不同的对象；因此，它可能无法区分病人和床。为了克服这一缺点，诺登斯医疗公司建议使用最先进的人工智能算法来赋予传感器“智能”能力，使其能够学习区分不同的检测对象。这可以通过识别心跳的周期性节奏或一个人摆动的手臂或腿来完成。Nodens Medical将开发一种‘智能’的人工智能支持的MmWave RTLS解决方案，该解决方案将：1.跟踪患者、工作人员和设备，以确定工作人员是否探望过患者或跟踪感染的传播；2.监测跌倒检测；3.跟踪生命体征。该设备将具有市场领先的精确度和‘即插即用’功能，易于安装，帮助NHS在降低成本的同时改善患者的预后。“