Intelligent Wearable Sensors for Predictive Patient Monitoring

Lead Research Organisation: University of Oxford
Department Name: Engineering Science

Abstract

There is an urgent, unmet need for reliable, continuous patient monitoring, both in hospitalised patients, and in home settings. Delays in recognition of the changes in vital signs associated with physiological deterioration worsen outcomes and increase healthcare costs. In hospital wards patients are mainly mobile and unmonitored by electronic systems. Manual observations of vital signs (heart rate, respiratory rate, blood pressure, temperature, oxygen saturation and level of consciousness) are made by a nurse every 4 - 12 hours, resulting in late recognition of deterioration. Over 40,000 patients deteriorate on UK wards each year until they are so unwell that they require admission to an Intensive Care Unit (ICU). Care is delayed in over a third of these patients, and the mortality rate for such emergency admissions is around one quarter.

Wearable monitoring systems are required to address the needs of ambulatory patients in hospital. However, no wearable systems have penetrated into clinical practice at scale, due to: (i) poor tolerance of existing wearable devices for vital-sign monitoring; (ii) a lack of robustness in the estimates of the vital signs that wearable sensors produce; (iii) very limited battery life that requires batteries to be re-charged at a rate that prevents their use on a large scale; and (iv) limited subsequent use of the data for clinical decision support.

We propose to develop a range of "intelligent" wearable sensors, with smart algorithms embedded within them, to overcome these limitations. Our disruptive sensor system will build on major EPSRC-funded grants that have demonstrated proof-of-concept: the "Hospital of the Future" Grand Challenge in Healthcare IT (2009-2013) and the "Centre of Excellence in Medical Engineering" (2009-2015, funded jointly by the Wellcome Trust).

Planned Impact

The proposed programme has the potential for very significant impact for patients in the hospital or in the home, by optimising patient management and improving outcomes. Patients who are deteriorating will be identified early, which will allow preventative action to be taken, avoiding serious escalation (such as unplanned admission to an ICU), and which will reduce the incidence of preventable morbidity, cardiac arrests, and death. Those hospital patients recovering faster than expected, and who are deemed to be sufficiently stable, can be discharged home earlier. Patient-specific care will be enabled using intelligent sensors that learn patient characteristics in real-time, thus improving patient outcomes by improving the efficacy of care provided. The translation of such systems into the "hospital-at-home" environment will provide benefit to patients with long-term conditions, such as chronic obstructive pulmonary disease (COPD) and heart failure - this will be achieved via predictive systems that allow clinicians to track patient condition without the false-alarm rate associated with existing systems, and which prevents existing systems providing benefit to existing patients.

The NHS as an organisation will benefit from the research because improved patient outcomes are associated with lower healthcare costs, as a result of shorter stays in hospital and fewer unplanned admission to (expensive) higher levels of care. Additionally, patients will be stratified according to risk of severity / deterioration, allowing improved use of clinical staffing resources. Clinicians will benefit by being able to interpret, for the first time, the very large and heterogeneous datasets that are available for their patients - enabled by robust, probabilistic tools created during the programme.

Companies in the commercial private sector will benefit from the research, where involvement of the industrial partners will allow rapid implementation of the techniques developed during the programme. Such companies have an interest in "intelligent healthcare" algorithms that can be integrated into existing healthcare IT products, which will add significant value and market differentiation. Additionally, the rapidly-growing market for wearable devices is currently focused only on consumers - the proposed work will extend this market to healthcare technologies, by exploiting the opportunities for large-scale innovation and clinical validation that exist in the programme. The UK economy will benefit by the possible creation of new spin-out activity based around the activity of the proposed work, in addition to the "first to market" advantage conferred on the industrial partners.

The scientific community, and the UK research base in particular, will benefit from developing capacity in an emerging field of global importance, and where the proposed project will train the next generation of researchers in intelligent healthcare sensing. The methodology developed within the proposed programme will be of translational benefit to other scientific disciplines, including other computational and mathematical sciences. Results from the research will feed into the Alan Turing Institute, where researchers involved across the UK will benefit from the development of large-scale sensing methodologies for data science.

The public will benefit via public-engagement activities run in collaboration with the Institution for Engineering & Technology (the world's largest multidisciplinary engineering professional institution), the George Institute for Global Health (which holds public-engagement conferences at its bases in Oxford, Sydney, Beijing, and New Delhi), the Royal Academy of Engineering (which supports Clifton's work via a Research Fellowship), and the NIHR Oxford Biomedical Research Centre (which holds regular public outreach events).

Publications

10 25 50

publication icon
Birrenkott DA (2018) A Robust Fusion Model for Estimating Respiratory Rate From Photoplethysmography and Electrocardiography. in IEEE transactions on bio-medical engineering

publication icon
Birrenkott DA (2016) Robust estimation of respiratory rate via ECG- and PPG-derived respiratory quality indices. in Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference

publication icon
Charlton PH (2018) Breathing Rate Estimation From the Electrocardiogram and Photoplethysmogram: A Review. in IEEE reviews in biomedical engineering

 
Description Failure to recognize and act on signs of worsening acute illness in acute hospital wards is a common problem that was recognized over a decade ago. Early Warning Systems (EWS) have pre-determined 'calling criteria' protocols initiating medical escalation of deteriorating patients. These protocols are limited by their dependence on correct frequency of vital-sign measurements. Continuous vital-sign monitoring may address this by identifying changes in patient condition earlier than intermittent measurements, and by releasing nursing time. Wearable monitors potentially provide benefits to patients including improved mobility and comfort over traditional 'bedside' systems. However, there is little data to support their use in hospital populations. There is a clear need to investigate integrating robust wearable technology into early warning systems to improve management of deteriorating patients. We have shown that 90% of the escalations from the general ward to the Intensive Care Unit can be identified early through the continuous monitoring and integration of three vital signs (heart rate, breathing rate and oxygen saturation) using our robust wearable technology system.

Our programme of research has examined the effect of continuous vital-sign monitoring on the management of clinical deterioration in emergency surgical patients. An initial development phase, which included human factors-based usability testing, has optimized the ambulatory system for vital-sign monitoring. The optimized system, integrated with current practice, is now being tested on wards in the John Radcliffe Hospital in Oxford.
Exploitation Route We now have a solution in which high-risk patients, who are continuously monitored with our wearable technology system, can be nursed with the same frequency of observations as low-risk patients. The intellectual property underpinning this solution should be of interest to patient monitoring companies.
Sectors Healthcare

 
Description Our findings were used in discussions with Sensyne Health in late 2019 on how to design a patient monitoring system which incorporates wearables. The commercial advantage is that the same frequency of observations for low-risk patients can be used for high-risk patients with wearables, and hence it is possible to manage a greater number of patients with the same number of nurses. The success of our wearables system for monitoring COVID patients on isolation wards (see below) has led to these discussions with Sensyne Health re-starting. In March 2020, our system was adapted for use on the patient isolation wards in the John Radcliffe Hospital in response to the outbreak of the COVID-19 pandemic. In the early stages of the pandemic, there was a fear that the lack of knowledge about the dynamics of virus transmission and initial shortages of personal protective equipment, could lead to an increased spread of the infection in hospital clinical staff. Continuous vital-sign monitoring enables the identification of the rapid desaturation caused by the SARS-CoV-2 virus, followed by intervention with corrective treatment at the earliest opportunity (e.g. via additional oxygen therapy). However, the vital-sign monitoring tools used in isolation wards, bedside monitors, are not ideal for ambulatory COVID-19 patients. Furthermore, point-of-care devices for nursing observations require nursing staff to be present in the patient isolation rooms to take accurate vital-sign measurements (e.g. body temperature), increasing the risk of spreading the infection. It became clear at the end of February 2020 that the technology and software developed for this research project could be adapted for the isolation ward for COVID-19 patients (John Warin ward). For those COVID-19 patients with no clinical requirement to be managed on a ventilator, or stepping down from intensive care, it is important for their recovery that they should continue to remain ambulatory, and so the wearable technology is ideal. The research team from the Institute of Biomedical Engineering worked tirelessly throughout the first three weeks of March to ensure that the modified wearable monitoring system could be deployed soon after the opening of the isolation ward. The system went live with its first four ambulatory patients on the John Warin ward on Monday 23rd March, the day that the first national lockdown started. The processed vital-sign data is made available to the nursing staff outside the isolation rooms using the hospital wi-fi, and is displayed on a dashboard which allows the physiological status of the patients to be tracked in real-time. Our wearables system uses web-based architecture and protocols (HTTP and WebSockets) to transmit the vital-sign data in real time from BLE and Wi-Fi enabled Android tablet devices, operating as patient data collection devices by the bedside in the isolation rooms, into a clinician dashboard interface available remotely via any modern web-browser. Fault-tolerant software strategies are used to reconnect the wearables automatically, during intermittent BLE disconnections, avoiding the need for nurses to enter the isolation ward to re-set the patient monitoring equipment. The remote dashboard also displays the vital-sign observations made by the nurses and entered by a separate system into the hospital electronic Track & Trigger record (eT&T) via HL7. The latter were displayed alongside the continuous wearable data, allowing nurses to review both sources of vital-sign data in one consistent augmented e-T&T screen. System usage was found to follow the two main COVID-19 wave trends, with half of the patients on the isolation ward on the wearable system during the peak of hospital admissions in both of the main waves (April-July 2020 and December 2020-March 2021). During the first wave, patients were monitored for a median of 31.5 [8.8, 75.4] hours, representing 88.1 [62.5, 94.5]% of the median time they were registered in the system. By 1st March 2021, 145 patients had been monitored for an overall 300 patient-days. Sarah Vollam, a Critical Care Nurse Researcher, says:"The biomedical engineers did an amazing job of making the wearables system for the COVID ward user-friendly and intuitive, which is key to implementing new technology in a stressful environment. The nurses were very keen to use it to enhance not only their patients' safety but also their own, by limiting their exposure to the virus".
First Year Of Impact 2020
Sector Healthcare
Impact Types Societal

 
Description Partnership with Drayson Technologies to deploy digital health technologies for the NHS 
Organisation Sensyne Health
Country United Kingdom 
Sector Private 
PI Contribution Three new digital health products, developed in my research group at the Institute of Biomedical Engineering in collaboration with clinical colleagues in the Oxford University Hospitals (OUH) NHS Foundation Trust over the past seven years, promise significant improvements in patient health outcomes and reduced costs for the NHS. These products have been exclusively licensed to Drayson Technologies. The three new digital health products are SEND, GDm-health and EDGE-COPD (see http://www.eng.ox.ac.uk/about/news/new-digital-health-products-developed-at-the-ibme-to-be-commercialised-by-drayson-technologies). The product which has come out of the research funded by this grant is SEND: a system for vital-sign observations in hospital patients (blood pressure, heart rate, breathing rate, oxygen saturation and temperature) and risk score estimation, which has enhanced the clinical care of over 80,000 patients over the past two years. As well as improving individual patient care, the SEND system allows information about patients to be shared between different wards. This ensures quicker decision-making and allows cross-linking with other patient data. On many wards, patients are now prioritised during the ward round according to their latest risk score. Our research team developed the risk estimation algorithm (see http://www.sciencedirect.com/science/article/pii/S030095721100195X, also known as an early warning score, and designed the original prototype for the system now deployed on all adult wards in the four acute hospitals in Oxfordshire.
Collaborator Contribution They will be productionising the SEND system and deploying in NHS Trusts beyond Oxfordshire.
Impact None yet
Start Year 2017
 
Title Wearable system for real-time vital-sign monitoring of in-hospital patients 
Description In March 2020, our wearables system was adapted for use on the patient isolation wards in the John Radcliffe Hospital in Oxford in response to the outbreak of the COVID-19 pandemic. In the early stages of the pandemic, there was a fear that the lack of knowledge about the dynamics of virus transmission and initial shortages of personal protective equipment, could lead to an increased spread of the infection in hospital clinical staff. Continuous vital-sign monitoring enables the identification of the rapid desaturation caused by the SARS-CoV-2 virus, followed by intervention with corrective treatment at the earliest opportunity (e.g. via additional oxygen therapy). However, vital-sign monitoring tools used in isolation wards, bedside monitors, are not ideal for ambulatory COVID-19 patients. Furthermore, point-of-care devices require staff to be present in the patient isolation rooms to take accurate vital-sign measurements (e.g. body temperature), increasing the risk of spreading the infection. It became clear at the end of February 2020 that the technology and software developed for this research project could be adapted for the isolation ward for COVID-19 patients (John Warin ward). For those COVID-19 patients with no clinical requirement to be managed on a ventilator, or stepping down from intensive care, it is important for their recovery that they should continue to remain ambulatory, and so the wearable technology is ideal. The research team from the Institute of Biomedical Engineering worked tirelessly throughout the first three weeks of March to ensure that the modified wearable monitoring system could be deployed soon after the opening of the isolation ward. The system went live with its first four ambulatory patients on the John Warin ward on Monday 23rd March, the day that the first national lockdown started. The processed vital-sign data is made available to the nursing staff outside the isolation rooms using the hospital wi-fi, and is displayed on a dashboard which allows the physiological status of the patients to be tracked in real-time. Our wearables system used web-based architecture and protocols (HTTP and WebSockets) to transmit the vital-sign data in real time from BLE and Wi-Fi enabled Android tablet devices, operating as patient data collection devices by the bedside in the isolation rooms, into a clinician dashboard interface available remotely via any modern web-browser. Fault-tolerant software strategies were used to reconnect the wearables automatically, during intermittent BLE disconnections, avoiding the need for nurses to enter the isolation ward to re-set the patient monitoring equipment. The remote dashboard also displayed the vital-sign observations made by the nurses and entered by a separate system into the hospital electronic Track & Trigger record (eT&T) via HL7. The latter were displayed alongside the continuous wearable data, allowing nurses to review both sources of vital-sign data in one consistent augmented e-T&T screen. System usage was found to follow the two main COVID-19 wave trends, with half of the patients on the isolation ward being monitored by our system during the peak of hospital admissions in both of the main waves of the pandemic (April-July 2020 and December 2020-March 2021). During the first wave, patients were monitored for a median of 31.5 [8.8, 75.4] hours, representing 88.1 [62.5, 94.5]% of the median time they were registered in the system. By 1st March 2021, 145 patients had been monitored for an overall 300 patient-days. 
Type Diagnostic Tool - Non-Imaging
Current Stage Of Development Early clinical assessment
Year Development Stage Completed 2020
Development Status Actively seeking support
Impact N.A. 
URL http://N.A.