SPHERE - A Sensor Platform for HEalthcare in a Residential Environment (IRC Next Steps)
Lead Research Organisation:
University of Bristol
Department Name: Electrical and Electronic Engineering
Abstract
The UK currently spends 70% of its entire health and social care budget on long term ("chronic") health conditions. These include diabetes, dementia, obesity, depression, COPD, arthritis, hypertension and asthma.
We need to be better at:
-- Understanding the cause of these illnesses
-- Helping a person to avoid developing them
-- Creating new treatments
-- Helping the patient self-manage their conditions
All these require working with a patient over months or years, outside of a traditional hospital environment. In a very real way, we need healthcare to go where the patient goes; the single place that most people spend most of their time is their home. Consequently, SPHERE project is seeking to develop non-intrusive home-based technologies for measuring health related behaviours at home over long periods of time.
The requirements for these technologies are:
-- They should require little or no action from the patient, since our daily lives are busy; being ill is distressing and time-consuming; and when the benefit may take months or years to achieve, there is often not much day to day motivation to be bothered with measurements or devices.
-- They should work reliably in the home; a home is not a hospital or a laboratory - it is smaller, full of furniture, pets and people, often not brightly-lit and often challenging to get wireless network coverage everywhere. This poses lots of problems for researchers.
-- They should be acceptable; bringing healthcare home with us doesn't mean we want to turn our homes into hospital and it definitely doesn't mean we want people spying on us!
Since 2013 this has been the SPHERE vision and we have worked with scientists, doctors, engineers and more than 200 members of the public to achieve the project's initial goal of creating a cheap sensor system that can be installed in a home. More than 30 people have had the experience of living with the sensors over periods from days to months and, by the end 2017 we expect more than 200 people will have had SPHERE sensors in their own home, in many cases for months.
Although the first-generation system was only completed in late 2016 and at the time of writing is still under test in the first "pilot" homes, the system is already moving into real patient applications - we are applying for ethical permission from the NHS to use SPHERE for patients recovering from surgery. Later in 2017 we will be applying for ethical permission to use SPHERE with a group of dementia patients.
The initial testing of the sensor system has gone well but, especially as we start to think about large scale use of the SPHERE system across potentially hundreds or thousands of people, the team have learnt a lot from the early pilots and have some priorities for significant improvements:
1. The SPHERE video system needs to be better at evaluating the quality of someone's movement, such as getting out of a chair, even when the view of the person is blocked by items of furniture. Evaluating quality of movement is important in physical and mental health conditions.
2. The SPHERE wristband lasts for over a month on a single charge, however we want to remove as far as possible the need to charge it at all, because the more ill someone is, the less likely they are to do this.
3. Digital data gathered from sensors needs to be turned into understanding for doctors; this is especially difficult in a home environment because every home and every household is different.
These are major research issues and will be the focus of the technology parts of the SPHERE programme, while the clinical parts move forward with patient populations.
The NHS itself has recently said: "if the UK fails to get serious about prevention then recent progress in healthy life expectancies will stall, health inequalities will widen, and our ability to fund beneficial new treatments will be crowded-out by the need to spend billions of pounds on wholly avoidable illness."
We need to be better at:
-- Understanding the cause of these illnesses
-- Helping a person to avoid developing them
-- Creating new treatments
-- Helping the patient self-manage their conditions
All these require working with a patient over months or years, outside of a traditional hospital environment. In a very real way, we need healthcare to go where the patient goes; the single place that most people spend most of their time is their home. Consequently, SPHERE project is seeking to develop non-intrusive home-based technologies for measuring health related behaviours at home over long periods of time.
The requirements for these technologies are:
-- They should require little or no action from the patient, since our daily lives are busy; being ill is distressing and time-consuming; and when the benefit may take months or years to achieve, there is often not much day to day motivation to be bothered with measurements or devices.
-- They should work reliably in the home; a home is not a hospital or a laboratory - it is smaller, full of furniture, pets and people, often not brightly-lit and often challenging to get wireless network coverage everywhere. This poses lots of problems for researchers.
-- They should be acceptable; bringing healthcare home with us doesn't mean we want to turn our homes into hospital and it definitely doesn't mean we want people spying on us!
Since 2013 this has been the SPHERE vision and we have worked with scientists, doctors, engineers and more than 200 members of the public to achieve the project's initial goal of creating a cheap sensor system that can be installed in a home. More than 30 people have had the experience of living with the sensors over periods from days to months and, by the end 2017 we expect more than 200 people will have had SPHERE sensors in their own home, in many cases for months.
Although the first-generation system was only completed in late 2016 and at the time of writing is still under test in the first "pilot" homes, the system is already moving into real patient applications - we are applying for ethical permission from the NHS to use SPHERE for patients recovering from surgery. Later in 2017 we will be applying for ethical permission to use SPHERE with a group of dementia patients.
The initial testing of the sensor system has gone well but, especially as we start to think about large scale use of the SPHERE system across potentially hundreds or thousands of people, the team have learnt a lot from the early pilots and have some priorities for significant improvements:
1. The SPHERE video system needs to be better at evaluating the quality of someone's movement, such as getting out of a chair, even when the view of the person is blocked by items of furniture. Evaluating quality of movement is important in physical and mental health conditions.
2. The SPHERE wristband lasts for over a month on a single charge, however we want to remove as far as possible the need to charge it at all, because the more ill someone is, the less likely they are to do this.
3. Digital data gathered from sensors needs to be turned into understanding for doctors; this is especially difficult in a home environment because every home and every household is different.
These are major research issues and will be the focus of the technology parts of the SPHERE programme, while the clinical parts move forward with patient populations.
The NHS itself has recently said: "if the UK fails to get serious about prevention then recent progress in healthy life expectancies will stall, health inequalities will widen, and our ability to fund beneficial new treatments will be crowded-out by the need to spend billions of pounds on wholly avoidable illness."
Planned Impact
Like many developed nations, the UK faces huge challenges dealing with long term health conditions such as diabetes, dementia, depression, COPD, arthritis and asthma. Whether seeking to understand the mechanisms, trying to avoid onset, creating new therapies, or supporting self-care of these conditions, we need non-intrusive technologies able to capture data on causes, symptoms and exacerbations over long periods of time.
SPHERE has been developed in partnership with health professionals, specifically with the intention of producing a sensor system that can be used by the NHS. Early SPHERE work is already funded with clinical cohorts e.g. to characterise recovery after Orthopaedic (Bristol) and Cardiovascular (Sheffield) surgery. There are on-going discussions with leading centres on subjects such as personalised behaviour change programmes, physical activity interventions in patients with COPD and new forms of free-living gait analysis.
Local government has a statutory responsibility for public health and SPHERE has close engagement with Bristol City Council, as described in its letter of support.
SPHERE was designed from the outset as a tool for clinical research, helping us understand what causes these illnesses and develop new treatments. SPHERE is already collaborating with many health researchers but as it now demonstrates its applicability through its 100 Homes trial, it will increasingly be working for impact in partnership with major UK institutes (UK Biobank, the Dementias Research Institute, Inst. for Biomedical Informatics) and continuing discussions with GSK around digital clinical trials.
SPHERE has a stated objective to engage with policy-makers e.g. with Public Health England (the Director spoke at their annual conference) and one of its Directors sits on SPHERE's advisory board.
Many health issues are global concerns, and indeed the SPHERE director has been invited to speak at the British Embassy in Tokyo in Feb 2017 to an audience of Japanese policy makers and academics on the subject of dementia (
countries such as Japan and Singapore have the world's "oldest" populations).
With Europe's 2nd largest medical technology industry, UK companies, large and small, are well-placed to address this need. SPHERE has met many SMEs (including Cascoda, Cascade3d, Pumpco, Folium Optics and Loc8tor) to assist them in their approach to this market. SPHERE will partner with organisations such as the Digital Catapult (see their letter of support) to help many more UK companies understand how to address those needs with new technologies, how to develop more reliable in-home wireless networks, develop analytics capabilities for domestic data, etc.
SPHERE is also developing relations with corporates including IBM, Toshiba, BT, Sony, TI, Dyson, BT & Jaguar Land-Rover. UK employers such as Loc8tor (see letter), Janssen Healthcare Innovation, McLaren Technology Group, IBM & Toshiba all have on-going research or stated intentions for collaborative research with SPHERE.
SPHERE will also work with standards bodies such as IETF (see letter from the 6TiSCH working group chair) to influence the evolution of wireless network standards for health applications. While the potential health benefit, the market and the industrial potential are all quite apparent, the adoption of pervasive in-home technology for health delivery runs the risk of creating a range of new issues, including around security, trust and privacy. By conducting its research in close partnership with the public (SPHERE was 1 of 3 finalists in the UK Public Engagement Awards), SPHERE will seek to disseminate the issues to the public in an open way - this indeed is one of the fundamental purposes and responsibilities of University research. Furthermore, within the programme proposed herein, SPHERE will particularly address the extent to which some populations might be differentially advantaged or disadvantaged by this type of health delivery
SPHERE has been developed in partnership with health professionals, specifically with the intention of producing a sensor system that can be used by the NHS. Early SPHERE work is already funded with clinical cohorts e.g. to characterise recovery after Orthopaedic (Bristol) and Cardiovascular (Sheffield) surgery. There are on-going discussions with leading centres on subjects such as personalised behaviour change programmes, physical activity interventions in patients with COPD and new forms of free-living gait analysis.
Local government has a statutory responsibility for public health and SPHERE has close engagement with Bristol City Council, as described in its letter of support.
SPHERE was designed from the outset as a tool for clinical research, helping us understand what causes these illnesses and develop new treatments. SPHERE is already collaborating with many health researchers but as it now demonstrates its applicability through its 100 Homes trial, it will increasingly be working for impact in partnership with major UK institutes (UK Biobank, the Dementias Research Institute, Inst. for Biomedical Informatics) and continuing discussions with GSK around digital clinical trials.
SPHERE has a stated objective to engage with policy-makers e.g. with Public Health England (the Director spoke at their annual conference) and one of its Directors sits on SPHERE's advisory board.
Many health issues are global concerns, and indeed the SPHERE director has been invited to speak at the British Embassy in Tokyo in Feb 2017 to an audience of Japanese policy makers and academics on the subject of dementia (
countries such as Japan and Singapore have the world's "oldest" populations).
With Europe's 2nd largest medical technology industry, UK companies, large and small, are well-placed to address this need. SPHERE has met many SMEs (including Cascoda, Cascade3d, Pumpco, Folium Optics and Loc8tor) to assist them in their approach to this market. SPHERE will partner with organisations such as the Digital Catapult (see their letter of support) to help many more UK companies understand how to address those needs with new technologies, how to develop more reliable in-home wireless networks, develop analytics capabilities for domestic data, etc.
SPHERE is also developing relations with corporates including IBM, Toshiba, BT, Sony, TI, Dyson, BT & Jaguar Land-Rover. UK employers such as Loc8tor (see letter), Janssen Healthcare Innovation, McLaren Technology Group, IBM & Toshiba all have on-going research or stated intentions for collaborative research with SPHERE.
SPHERE will also work with standards bodies such as IETF (see letter from the 6TiSCH working group chair) to influence the evolution of wireless network standards for health applications. While the potential health benefit, the market and the industrial potential are all quite apparent, the adoption of pervasive in-home technology for health delivery runs the risk of creating a range of new issues, including around security, trust and privacy. By conducting its research in close partnership with the public (SPHERE was 1 of 3 finalists in the UK Public Engagement Awards), SPHERE will seek to disseminate the issues to the public in an open way - this indeed is one of the fundamental purposes and responsibilities of University research. Furthermore, within the programme proposed herein, SPHERE will particularly address the extent to which some populations might be differentially advantaged or disadvantaged by this type of health delivery
Publications
Ayobi A
(2022)
Digital Mental Health and Social Connectedness Experiences of Women from Refugee Backgrounds
in Proceedings of the ACM on Human-Computer Interaction
Ayobi A
(2020)
Trackly
Bi H
(2021)
Human Activity Recognition Based on Dynamic Active Learning
in IEEE Journal of Biomedical and Health Informatics
Bi H
(2022)
An active semi-supervised deep learning model for human activity recognition
in Journal of Ambient Intelligence and Humanized Computing
Bi H
(2020)
Polarimetric SAR Image Semantic Segmentation With 3D Discrete Wavelet Transform and Markov Random Field
in IEEE Transactions on Image Processing
Burrows A
(2018)
User involvement in digital health: Working together to design smart home health technology
in Health Expectations
Description | A study with patients with Parkinson's Disease has captured in unique detail the lived experience of those with this incurable, debilitaling, illness. A particular focus is developing behavioural metrics that could characterise the response of a patient to medication. |
Exploitation Route | The behavioural metrics could be used by pharma companies to help trial new medications. |
Sectors | Healthcare |
Description | Advisor to the National Data Guardian |
Geographic Reach | National |
Policy Influence Type | Participation in a guidance/advisory committee |
Title | LoMAPP |
Description | LoMAPP: Localisation state as a MedicAtion Proxy in Parkinson's disease. This dataset is linked to a paper published through the conference 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (https://kdd.org/kdd2023/). Title of paper: Multimodal Indoor Localisation in Parkinson's Disease for Detecting Medication Use: Observational Pilot Study in a Free-Living Setting. Authors: Ferdian Jovan, Catherine Morgan, Ryan McConville, Emma L. Tonkin, Ian Craddock, Alan Whone. The dataset creation is described in this paper. |
Type Of Material | Database/Collection of data |
Year Produced | 2023 |
Provided To Others? | Yes |
URL | https://data.bris.ac.uk/data/dataset/3s3ab1sgszqy52escorp07ss8j/ |
Title | REMAP Controlled dataset |
Description | Parkinson's disease (PD) is a neurodegenerative disorder characterised by motor symptoms, such as gait dysfunction and postural instability. Technological tools to continuously monitor outcomes could capture the hour-by-hour symptom fluctuations of PD. Development of such tools is hampered by the lack of labelled datasets from home settings. To this end, we propose REMAP, a human rater-labelled dataset of REal-world Mobility Activities in Parkinson's disease including people with and without PD doing sit-to-stand transitions and turns in gait while living in a home setting. These discrete activities are captured from free-living (unobserved, unstructured) and during clinical assessments. The PD participants withheld their dopaminergic medications for a time (causing increased symptoms), so their events are labelled as being "on" or "off" medications. We include accelerometry from wrist-worn wearables and skeleton pose camera data. We present this controlled dataset available on application where there is more refined data (this is a sister dataset to the open dataset, REMAP Open, where the data is coarsened for anonymisation). |
Type Of Material | Database/Collection of data |
Year Produced | 2023 |
Provided To Others? | Yes |
URL | https://data.bris.ac.uk/data/dataset/2o94rzjooyzf42w850dqg0spfh/ |
Title | REMAP Open dataset |
Description | Parkinson's disease (PD) is a neurodegenerative disorder characterised by motor symptoms, such as gait dysfunction and postural instability. Technological tools to continuously monitor outcomes could capture the hour-by-hour symptom fluctuations of PD. Development of such tools is hampered by the lack of labelled datasets from home settings. To this end, we propose REMAP, a human rater-labelled dataset of REal-world Mobility Activities in Parkinson's disease including people with and without PD doing sit-to-stand transitions and turns in gait while living in a home setting. These discrete activities are captured from free-living (unobserved, unstructured) and during clinical assessments. The PD participants withheld their dopaminergic medications for a time (causing increased symptoms), so their events are labelled as being "on" or "off" medications. We include skeleton pose camera data. We present this open dataset, where the skeleton pose data is coarsened for anonymisation. We illustrate use-case code measuring sit-to-stand duration, tested on both coarsened and refined data. This is a sister dataset so our controlled dataset (available on application) where there is more refined skeleton pose data and accelerometry. |
Type Of Material | Database/Collection of data |
Year Produced | 2023 |
Provided To Others? | Yes |
URL | https://data.bris.ac.uk/data/dataset/21h9f9e30v9cl2fapjggz4q1x7/ |
Title | SPHERE house scripted dataset: A multi-sensor dataset with annotated activities of daily living recorded in a residential setting |
Description | This dataset expands the original Sphere Challenge dataset with access to a full set of both training and testing data. This dataset is well-adapted for use to predict aspects of the activities of residents within a smart home based only on observed sensor data. An accompanying data descriptor is under submission to Nature Scientific Data. |
Type Of Material | Database/Collection of data |
Year Produced | 2020 |
Provided To Others? | Yes |
Impact | A multi-sensor dataset with annotated activities of daily living recorded in a residential setting is an extended version of an older dataset that has had significant impact including a data challenge and use in teaching and learning. There is also a Nature data descriptor that came back with major corrections, so we will find out once we resubmit whether that succeeds in getting published. |
URL | https://data.bris.ac.uk/data/dataset/1dcxespcsafm02fba4bckx0ymt/ |
Title | SPHERE house scripted dataset: A multi-sensor dataset with annotated activities of daily living recorded in a residential setting v2.0 |
Description | This dataset expands the original Sphere Challenge dataset with access to a full set of both training and testing data. This dataset is well-adapted for use to predict aspects of the activities of residents within a smart home based only on observed sensor data. An accompanying data descriptor is under submission to Nature Scientific Data. An updated version was deposited in November 2022 with extensive additional data, as well as being reorganised to facilitate accessibility and reusability. This dataset is available under the CC-BY licence. Earlier version of this dataset available at: https://doi.org/10.5523/bris.1dcxespcsafm02fba4bckx0ymt |
Type Of Material | Database/Collection of data |
Year Produced | 2022 |
Provided To Others? | Yes |
Impact | A subset of this dataset was previously released to facilitate a machine learning competition in association with the 2016 European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, which attracted 580 participants worldwide. Since then it has been in use for teaching and learning and in applied machine learning for purposes including the development and validation of activity recognition algorithms. The Nature Data release makes available the full, carefully curated dataset as a general resource intended to benefit healthcare monitoring and activity recognition research. This dataset is one of the most popular datasets released on the University of Bristol's data repository by page views [http://www.bristol.ac.uk/staff/researchers/data/metrics/] |
URL | https://data.bris.ac.uk/data/dataset/2h0wyctxrd69j2oqccsi45hy1p/ |
Title | Sphere House Multi-wearable |
Description | A dataset containing data from the Wearable 3 sensor developed by SPHERE. This dataset contains two sessions in which participants complete a simple set of activities of daily living, wearing multiple wearables (1-2 wearables on each wrist). It uses a methodology described in a recent paper, "Towards a Methodology for Acceptance Testing and Validation of Monitoring Bodyworn Devices" |
Type Of Material | Database/Collection of data |
Year Produced | 2020 |
Provided To Others? | Yes |
Impact | SPHERE house multi-wearable has been widely used for teaching purposes within the university of Bristol. It was also reported in a workshop (ARDUOUS 2020, hosted within PerCom) with an accompanying publication https://ieeexplore.ieee.org/document/9156257 |
URL | https://data.bris.ac.uk/data/dataset/pvn2rvkfirba2rz0ul145q18a/ |
Description | Collaboration with IXICO ltd |
Organisation | Ixico |
Country | United Kingdom |
Sector | Private |
PI Contribution | Collaborated on a research protocol and ethics application for a study involving sensor technology and a cohort of patients with Parkinson's Disease. Data from the study was provided to IXICO. |
Collaborator Contribution | Collaborated on a research protocol and ethics application for a study involving sensor technology and a cohort of patients with Parkinson's Disease. The collaboration involved funding from IXICO and donation of equipment. |
Impact | Published paper: An Automatic Gait Analysis Pipeline for Wearable Sensors: A Pilot Study in Parkinson's Disease, LR Peraza, KM Kinnunen, R McNaney, IJ Craddock, AL Whone et al, Sensors vol 21 (24), 8286. Multidisciplinary: Health Science, Computer Science, Electronic Engineering. |
Start Year | 2019 |
Description | Collaborative Paper with Newcastle University |
Organisation | Newcastle University |
Department | School of Computing Science |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | Jointly developed a protocol and wrote a protocol paper. |
Collaborator Contribution | Jointly developed a protocol and wrote a protocol paper. Newcastle's contribution being predominantly technical. |
Impact | 10.1136/bmjopen-2020-041303 10.3233/JPD-191781 |
Start Year | 2022 |
Description | We The Curious 'Hidden Health' Project |
Organisation | We The Curious |
Country | United Kingdom |
Sector | Charity/Non Profit |
PI Contribution | SPHERE funded the development of an outreach programme with local students. |
Collaborator Contribution | WeTheCurious co-designed the programme. |
Impact | Impacts were limited due to Covid limiting the contact we could have with young people and their schools. 3 young people joined the programme which resulted in an exhibit at WeTheCurious (again, limited by venue closures during the pandemic), |
Start Year | 2020 |
Description | EPSRC Healthcare Technology Team Visit to Bristol |
Form Of Engagement Activity | Participation in an open day or visit at my research institution |
Part Of Official Scheme? | No |
Geographic Reach | Regional |
Primary Audience | Other audiences |
Results and Impact | The EPSRC's Technology team were hosted in Bristol by the SPHERE Project, with a morning poster session featuring research projects of the Digital Health and Care CDT's PhD students, an afternoon tour of the SPHERE Lab House and a presentation to and Q&A with academics and researchers from across Bristol University in the afternoon. |
Year(s) Of Engagement Activity | 2022 |
Description | Feature in Rory Cellan Jones blog |
Form Of Engagement Activity | Engagement focused website, blog or social media channel |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Public/other audiences |
Results and Impact | Ex BBC technology presenter and patient with Parkinsons Disease Rory Cellan Jones visited the SPHERE house, interviewed staff and later posted on his blog. |
Year(s) Of Engagement Activity | 2021 |
Description | ISPOR Webinar: Achieving Fit for Purpose Data from Wearables for Age-Related Diseases |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Other audiences |
Results and Impact | Guest lecture, advertised internationally. Audience was composed of global regulators and payers in the academia and pharmaceutical/biotech sectors. |
Year(s) Of Engagement Activity | 2023 |
Description | Royal College of Surgeons of England Future of Surgery Innovation Hub Round Table Event |
Form Of Engagement Activity | A formal working group, expert panel or dialogue |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Professional Practitioners |
Results and Impact | Shaping the creation of an innovation network for NHS surgeons. |
Year(s) Of Engagement Activity | 2022 |
Description | SPHERE Celebration Event |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Policymakers/politicians |
Results and Impact | Approximately 100 people attended Bristol University's Wills Memorial Building for a celebration event marking 10 years of the SPHERE project. Guests included the EPSRC technology team, the University's Vice Chancellor, members of SPHERE's public patient groups and advisory board, and digital health PhD and MSc students from the University of Bristol. |
Year(s) Of Engagement Activity | 2022 |