ORUK-VA-0006

Section 1 - Basic information about you and your application:

Title of research project
COMPASS (Continuous Monitoring and Patient Assessment)

Duration
18

Start date
16/10/2023

Profession
Entrepreneur

Your current job title/position
Director

Have you previously received funding from Orthopaedic Research UK or Versus Arthritis
No

Please provide project ref number for previous funding from Orthopaedic Research UK


Please provide project ref number for previous funding from Versus Arthritis


Did you attend the AI in Orthopaedic Conference organised by Orthopaedic Research UK in 2022?
No

Did you attend the Microsoft AI training courses organised by Orthopaedic Research UK?
Yes

What other AI training courses have you previously attended?
I completed my PhD in Computer Engineering at the University of Southampton which includes AI/ML trainings and Data science applied to the project. I also completed Microsoft Azure trainings associated to deployment of AI on the Azure platform. I had experience rolling AI solutions on Azure as part of my PhD and professional experience.

Are you an early-career researcher (ECR)? (definition of ECR)
yes

Section 2 - Lay summary

Lay summary:

The benefits of exercise and physical activity are well-known and documented but do not consider the stress caused on the joint and bone structure by mechanical load related to movements. Specifically, too little load might lead to osteoporosis (i.e., bone density grows lower than average) while too much might lead to osteoarthritis (e.g. microfractures). Measuring loads is currently achieved only through implants (e.g. knee replacement) or within specialized facilities (e.g., rehabilitation). Either of these stages requires to be seen by a physiotherapist and/or HCP introducing burden on the NHS and delays that might worsen the MSK conditions.

As of today, “healthy” load is not characterized and patients affected with MSK require physiotherapy consultations to obtain personalized exercise recommendations. However, mechanical loads are not continuously monitored outside of clinical setting, nor reported in an objective manner. So, the impact and fatigue caused by occupational, and activities performed routinely (e.g. going up and down from the stairs, standing on feet all day, etc..) can raise a barrier for individual to follow exercise recommendations. It is estimated that 19m people in the UK suffer from musculoskeletal problems resulting in ever-lengthening NHS waiting lists, and the loss of nearly 31 million working days. This represents an estimated cost of £7 billion a year to the UK economy and £5 billion spent each year in NHS treatment costs.

The ambition of this project is to support self-management and monitoring of individuals with MSK. Smart devices can be used to continuously monitor and define personalized “healthy” mechanical load. This is different than the physical activity guidelines for general population (e.g., 10k steps per day) which do not consider MSK symptoms (e.g., pain swelling) nor lower mobility (i.e., elderly, sedentary or obese). Smartphones and wearables are now broadly available to the MSK population and can be used to continuously estimate load baselines. Along recording of symptoms (e.g., pain, swelling) and medications, personalized exercise recommendations can be generated via AI for patients prior and/or in support to meeting with physiotherapist or HCP.

Section 3 - Purpose of research

Purpose of research:

The aim of this project is to use sensors in wearable and smartphone to provide definitive guidance on what constitutes a ‘good’ or ‘healthy’ load for individuals with MSK conditions. These devices create very large datasets which can be processed via AI to profile individual trends and generate recommendations based on mechanical loading.

The objectives are:

• Collect datasets of load from healthy and MSK population outside of clinical setting
• Define thresholds of load intensity (high, medium, low) and frequency (intermittent or constant)
• Define physio recommendations based on mechanical loads
• Model physical activity recommendations generated via AI
• Evaluate the feasibility to monitor adherence to physiotherapist recommendations via wearable

The key deliverables are:

• Datasets of load estimates generated by users from cohort of MSK patients
• Standardized load thresholds to classify MSK profiles from very active to sedentary
• Guidelines and recommendations of physical activity based on load estimates captured outside of clinical setting

Section 4 - Background to investigation

Background to investigation:

The effects of exercises and occupational activities require further studies, and in 2017, a consensus study was conducted to harmonize the classification methods. The group included OA and PA international professionals that agreed to use MET-min per week for studies measuring PA. The key recommendations also raised the need to measure the intensity and duration of the joint load (Gates et al., 2017), which at the time of this project are yet to be standardized.

The bone density and cartilage of the joints are critical musculoskeletal components to individual´s overall mobility and ability to take part in physical activities. The bone density depends on the load applied which is generally known as osteogenic loading. Variations in these loads occur in physical activities (standing, running, etc..) because of the impact of the lower limbs (hips, knees, feet) on the ground (known as ground force reaction).

The principle of bone adaptation is known as Wolff´s law and has been extensively reviewed through experimental and observational studies of bone changes (Ruff et al., 2006). Bone density and remodeling can be improved as a response to the mechanical caused by exercise, but it is essential to distinguish the different types of exercises (O’Brien, 2001). The reliability of accelerometers to measure the effects of loading and unloading generated by physical activity has been confirmed for squats movement (Bobbert, 2014) as well as with a football (Boyd et al., 2011) and a rugby team (McLean et al., 2018). Accelerometers have also been used to evaluate physical activity in musculoskeletal studies but outside the clinical environment, concluding that RA patients are more sedentary than control participants (Prioreschi et al., 2013). That correlation exists with their disease activity (Hernandez-Hernandez et al., 2014). A review of the techniques to measure weight bearing was conducted in 2003 (Hurkmans et al., 2003), but since then, accelerometers have been embedded in devices such as smartphones and wearables. They have become more broadly available to the general public.

The priority for patients is to manage their pain and reduce their medication. Thus instructions beyond prescriptions are expected to achieve their goals (Leach, 2018). Early survey results amongst people with arthritis (Gecht et al., 1996) have shown that physical activity is directly related to the MSK patient’s understanding of the benefits and ability to perform. Yet, studies have shown that even when the benefits are understood, the anxiety associated with pain remains a barrier for patients to stick to such programmes, and the adherence rate remains low (Vervloesem et al., 2012). According to a systematic review of 20 studies (Kirsten Jack, 2010), the barriers to treatment adherence in musculoskeletal physiotherapy are associated with motivational and psychological factors such as anxiety, depression, and social and family support. Low physical activity at baseline and worsening pain during exercise supplement the logistics challenges caused by work schedules, lack of time and financial constraints.

Current recommendations consider step count and MET-min as known unit of measure for a range of activities which has been well documented (Mancuso et al., 2007) (White et al., 2016). This project is an extension to the research conducted by the team to continuously collect, monitor and estimate mechanical loads using smartphones and smartwatches. The group conducted a study conducted to evaluate the correlation between load rate estimates obtained from wearables and force plate data using treadmills and found significant correlation (https://doi.org/10.1177/2055668320929). This further led to a PhD thesis assessing the validation of load rate magnitude algorithm and attempt to develop activity classifier (https://eprints.soton.ac.uk/422272/).

My PhD focused on developing a methodology using smartphone as primary source of data capture to continuously monitor physical activity and compare it to other wearables (Fitbit and smartwatches). Feedback from RA patients were captured in PPI group session and alongside the data gathering, also with RA patients. Further feedback from health clinicians were captured in poster presentation at BSR seminar and various presentations with NHS staff. The methodology has also been used in clinical study with OA patients (https://doi.org/10.1186/s12891-021-04409-z) in the context of clinical trial of knee injection. Beyond MSK, the methodology was used in the context of occupational study which concluded that existing classification do not consider the variations in load (DOI: 10.1016/j.rehab.2021.101619). The publication of these studies led to listing the platform as tool by the European Alliance of Associations for Rheumatology (EULAR – https://www.eular.org/) but further studies are required to be able to standardize the measure of load rate.

Section 5 - Plan of investigation

Plan of investigation:

In order to achieve the objectives listed, a multi-disciplinary approach will be taken.

Data will be collected from a variety of sources, including surveys, interviews, and observation. Specifically, data will continuously be collected from wearable (Infitex insole provided to patients) and smartphone sensors via App provided and loaded on patient´s phones. AI/ML algorithms will be developed to model physical activity recommendations.

The insights gained from wearable technology will be utilized to monitor adherence to physiotherapist recommendations and improve the overall quality of healthcare for individuals with musculoskeletal conditions.

The research plan will follow below steps:

• File and obtain appropriate ethics including with NHS and Southampton General Hospital R&D as patients will be recruited
• Research project to be presented to Southampton General Hospital MSK and Physiotherapists staff to raise awareness in monthly meeting
• Patients to be screened from Southampton General Hospital database with main inclusion criteria defined as patients diagnosed with MSK conditions currently seeking advice from Southampton General Hospital Physiotherapists
• Patients will be interviewed (and introduced to Patient Information Sheet), consented, and recruited at Southampton General Hospital Clinical Research Facility
• Patients will be provided with Infitex insole and the app to download on their phone with introduction to the protocol
• Load Data generated will be collected and processed remotely using Azure cloud based services. Train AI model of load and activity correlation.
• Survey with physio on physical activity recommendations. Train AI model to identify patients follow through on recommendations.
• Survey with physio on recommendations following visualization of load. Train AI model to generate recommendations.

Section 6 - Research environment and resources

Research environment and resources:

Prior to the start of the study, ethical approval will be obtained from the Southampton General Hospital Research Ethics Committee. All patients will be informed about the purpose and procedures of the research and provide written informed consent prior to their participation. The study will be conducted in accordance with the principles of Good Clinical Practice (GCP) as laid down by the International Conference on Harmonisation. Furthermore, all information collected will be treated in accordance with the Data Protection Act 2018.

Therefore the project requires support from Southampton General Hospital´s staff for patient screening and to ensure compliance of the research with NHS guidelines. The patients will be recruited at the Clinical research facility of the hospital but no specific lab space is required.

Patients will be provided with custom wearable device built by Infitex to measure the force load. The device can analyze the force load at various points of the foot while walking, running or jumping and has been developed to provide an accurate assessment of a patient’s gait pattern. The device also provides valuable feedback from a biomechanics perspective, allowing for an improved understanding of force load. These insoles will be used as reference point to assert that patient´s feet are in contact with the ground (also known as ground reaction force load).

Due to practical aspect to obtain the insole devices, the project aims at recruiting up to 100 patients. Patients will be expected to have a smartphone and provided with the details of an app to load on their phone. The app (Rapp/OApp) is listed as research tool in EULAR website.

No additional funding is expected for this project.

Section 7: Research impact

Who will benefit from this research?

This project is designed to improve patient outcomes by providing access to timely and accurate exercise information. It has the potential to significantly reduce patient wait times and access to recommendations, while physiotherapists could gain an insight into the patient experience and better understand their needs. The NHS could also benefit from increased efficiency and cost savings associated to population mobility.

How can your research be translated in real-life?

Wearable-based monitoring and recommendation systems allow for a person’s condition to be monitored in real-time, from anywhere in the world. This can help provide timely feedback to physiotherapists, leading to better outcomes for patients. Furthermore, such systems can be used by physiotherapists to provide remote management of their patients, allowing them to deliver personalized care even when time or distance is an issue. Finally, such systems could also provide a platform for physiotherapists and their patients to communicate with each other more efficiently, enabling them to work collaboratively towards recovery goals.

How will your research be beneficial for Orthopaedic Research UK, Versus Arthritis and their purpose?

This project aligns with the purpose of Orthopaedic Research UK and Versus Arthritis by providing an opportunity to explore an innovative way of tackling physical activity for individuals with MSK conditions. It will explore how data generated by wearables can be used to research and develop effective recommendations that can improve patient outcomes and reduce the burden on the NHS. Furthermore, it will strive to improve support for patients by providing a platform to understand their needs and how MSK conditions affect their day-to-day lives. By engaging with stakeholders both within and outside of the healthcare system, this project seeks to make tangible progress in improving the lives of those affected by MSK conditions.

Section 8: Outreach and engagement

Outreach and engagement

I plan to communicate my findings to the non-academic public through posters at the hospital and the website of the hospital. Additionally, I plan to participate to MSK patient and public involvements.
I will also work on publishing articles for popular publications in order to reach an even wider audience.

Section 9: Research budget

Requested funding from Orthopaedic Research UK and Versus Arthritis

University fees (if any)
£0

Salary
£49000

Consumables
£26000

Publications
£10000

Conference attendance
£10000

Other items
£

Total 'requested fund'
£95000



Other items




Other secured funds

Internal funding
£0

Partner (University)
£0

Partner (Commercial)
£0

Partner (Charity)
£0

Other sources
£0

Total 'other funds'
£0



List all the 'other sources' and explain how their funds are used to cover the costs of your research.

Section 10: Intellectual property and testing on animal

Is there an IP linked to this research?
No

Who owns and maintains this patent?


Does your research include procedures to be carried out on animals in the UK under the Animals (Scientific Procedures) Act?
No

If yes, have the following necessary approvals been given by:

The Home office(in relation to personal, project and establishment licences)?


Animal Welfare and Ethical Review Body?


Does your research involve the use of animals or animal tissue outside the UK?
No

Does the proposed research involve a protected species? (If yes, state which)



Does the proposed research involve genetically modified animals?


Include details of sample size calculations and statistical advice sought. Please use the ARRIVE guidelines when designing and describing your experiments.


There should be sufficient information to allow for a robust review of any applications involving animals. Further guidance is available from the National Centre for the Replacement, Refinement and Reduction of Animals in Research (NC3Rs), including an online experimental design assistant to guide researchers through the design of animal experiments.

Please provide details of any moderate or severe procedures


Why is animal use necessary, are there any other possible approaches?


Why is the species/model to be used the most appropriate?