PROTECT: Predictive approaches in managing long-term conditions: from remote monitoring data to digital biomarkers

Lead Research Organisation: Imperial College London
Department Name: Brain Sciences

Abstract

The cost of care for long-term conditions is up to 70% of total health and social care expenditure and account for 50% of GP appointments in the UK. Developing data-driven preventative and predictive measures will enhance the quality of care services and reduce the cost of managing long-term conditions.

Emerging technologies such as the Internet of Things and wearable devices provide new opportunities to collect continuous in-home monitoring data such as movement, daily activity, vital signs and sleep. This data provides new possibilities to monitor the progression of conditions, do risk assessments and predict adverse healthcare events.

This research will develop new information engineering and machine learning methods to integrate and analyse continuous in-home monitoring data and create new digital biomarkers for personalised and preventative care. It will create scientific advances in in-home monitoring technologies and machine learning applied to healthcare and, in particular, managing long-term conditions. It will develop clinically applicable and care informed solutions to utilise remote monitoring data to extract health insights and provide predictive analysis and healthcare risk assessments. The research will contribute to the ambitions outlined in the call to transform care and health at home and enable independence by producing affordable solutions that be applied across a range of long-term health conditions.

The research programme will create a digital platform to integrate environmental and physiological information from in-home observation and measurement technologies safely and securely. It will apply machine learning models and solutions to derive digital biomarkers that can be used to improve healthcare. The overarching aim is to use the system to extract relevant information from complex datasets to allows effective and timely health interventions. To achieve this, the team will build the software infrastructure that allows sensor information to be collected in a safe and secure, and privacy-aware way and create the fundamental building blocks for the analysis of digital biomarkers through the development of person-centred and clinically informed predictive approaches.

Publications

10 25 50
 
Title TIHM/UKDRI Dataset 
Description The dataset contains in-home sensory observation and measurement data from homes of people with dementia (n=170 homes). This includes high-resolution data from the participants' in-home movement, use of home appliances, daily activities, sleep, vital signs and daily wellbeing questionnaire. The cohort was part of the previous TIHM study. The study has continued as part of the UK DRI CR7T study since September 2019. The TIHM and UK DRI CR&T studies have collected over 40,0000 days of in-home observation and monitoring data and over 20,000 nights of sleep data collected using smart connected sensors. The dataset is currently used in our research to develop analytical models and digital biomarkers to predict adverse health conditions in dementia care. 
Type Of Material Database/Collection of data 
Year Produced 2020 
Provided To Others? No  
Impact We are using this dataset to develop digital biomarkers. The research is currently focused on identifying digital biomarkers that can be used for early detection of behavioural changes, infections, sleep disorders, risk of falls and social isolation. The current dataset and the previously collected dataset from the TIHM study are partially labelled and annotated for events such as infections, hospital admissions, hypertension, agitation and changes in the wellbeing of people living with dementia. The participants have given their consent for their data to be used for research and the study has obtained approval from an NHS ethics review panel. 
 
Title TIHM: An open dataset for remote healthcare monitoring in dementia 
Description Dementia is a progressive condition that affects cognitive and functional abilities. There is a need for reliable and continuous health monitoring of People Living with Dementia (PLWD) to improve their quality of life and support their independent living. Healthcare services often focus on addressing and treating already established health conditions that affect PLWD. Managing these conditions continuously can inform better decision-making earlier for higher-quality care management for PLWD. The Technology Integrated Health Management (TIHM) project developed a new digital platform to routinely collect longitudinal observational and measurement data within the home and apply machine learning and analytical models for the detection and prediction of adverse health events affecting the well-being of PLWD. This work describes the TIHM dataset collected during the second phase (i.e., feasibility study) of the TIHM project. The data was collected from homes of 56 PLWD and associated with events and clinical observations (daily activity, physiological monitoring, and labels for health-related conditions). The study recorded an average of 50 days of data per participant, totalling 2803 days. 
Type Of Material Database/Collection of data 
Year Produced 2023 
Provided To Others? Yes  
Impact This dataset has been used to develop several models and analysis in dementia care, including: [1] Palermo et al., https://arxiv.org/abs/2110.09868 [2] Huang et al., https://arxiv.org/abs/2210.01736 [3] Rezvani et al., https://doi.org/10.1002/alz.052181 [4] Enshaeifar et al., https://doi.org/10.1145/3366424.3383541 [5] Rezvani et al., https://doi.org/10.1109/TKDE.2019.2961097 [6] Fletcher-Lloyd et al., https://doi.org/10.1101/2022.10.25.22281467 [7] Serban et al., https://doi.org/10.1038/s41746-022-00697-4 
URL https://zenodo.org/record/7622128
 
Description Jawdrop Life Sciences Summit: Addressing Healthy Ageing, Pandemics, and AMR 
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 A panel discussion on how the White City life science cluster is making old age better
Year(s) Of Engagement Activity 2024
URL https://www.whitecityinnovationdistrict.org.uk/jawdrop-life-sciences-24-putting-resilience-at-the-he...
 
Description Keynote, the 7th Singapore International Neuro-Cognitive Symposium 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact A keynote at the 7th Singapore International Neuro-Cognitive Symposium. The talk focused on the applications of AI in community care for dementia.
Year(s) Of Engagement Activity 2024
 
Description Teaching and educational material on Machine Learning for Neuroscience 
Form Of Engagement Activity Engagement focused website, blog or social media channel
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact This is an online set of material on machine learning methods, with a discussion of their use in neuroscience. We will cover several aspects of machine learning, from classical methods to deep learning.
Year(s) Of Engagement Activity 2024
URL https://ml4ns.github.io