MICA: Application for a Mental Health Data Pathfinder award (Oxford)

Lead Research Organisation: University of Oxford

Abstract

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.

Technical Summary

Making data findable and interoperable, adding value to existing datasets and enriching data collection from person-centric and patient reported measures represents the most valuable opportunity to accelerate understanding and improved treatment of mental health conditions.
Building on our experience of leading national and international programmes enabling access, interoperability and reuse of data for mental health - including the platform enabling use of Electronic Medical Records for mental health across England (CRIS), the platform for data aggregation for dementia (DPUK) and the pan-European platform to enable data interoperability from cohort and real-world data (EMIF) we propose here a programme of work including
underpinning themes and four indicative projects. These projects will gain from the linkage we propose to establish between routine care data and primary, secondary, schools and cohort data and will use devices to collect data remotely ranging from patient reported outcome measures to cognition, sleep and behaviour. To these enormous volumes of data, including the full mental health medical records from 10k people we have already linked to UK Biobank and the generation of 10k proteomes linked to mental health records, plus the streaming data from connected devices, we will bring advanced analytics from the Mathematics Institute and Natural Language Processing from computer scientists using neural networks. All our work will be enveloped in a governance and ethics stream with patient and public involvement. We leverage more than £40m in existing grant funding plus new cash and in-kind contribution from industry partners of over £12m. The outcome will be an informatics supported platform for experimental
medicine, increased understanding and development of novel therapeutics for depression and a bold drive to determine the relationship between intervention and outcome in mental wellbeing in young people. These themes and proposed projects will work with our proposed partners to establish the preparatory work for a platform for Mental Health Data capitalising on the unique opportunities presented by the NHS, by strengths in mental health research and by outstanding
digital sciences in the UK and in Oxford.

Publications

10 25 50

publication icon
Morgan A (2019) Inflammatory biomarkers in Alzheimer's disease plasma in Alzheimer's & Dementia

 
Description Collaboration between Oxford MRC Pathfinder Colleagues and Oxfordshire County Council - collected and provided data (and gave advice) to the council on pupils' mental health from participating schools in the region
Geographic Reach Local/Municipal/Regional 
Policy Influence Type Gave evidence to a government review
Impact The local authority will contribute to future data collections and liaise with the MRC Pathfinder team to understand the mental health of pupils in Oxfordshire, and use the data and advice from experts in the field to guide policy. Report to be published this month (February 2020) on the Joint Strategic Needs Assessment website, summarising the mental health of students in Oxfordshire
URL http://insight.oxfordshire.gov.uk/cms/joint-strategic-needs-assessment
 
Description AI for the discovery of new therapies in Parkinson's (JBCF)
Amount £200,000 (GBP)
Funding ID A2926 (Rosetrees) 
Organisation John Black Charitable Trust 
Sector Charity/Non Profit
Country United Kingdom
Start 03/2020 
End 04/2023
 
Description Alzheimer's gut microbiome project (NIH)
Amount $5,200,000 (USD)
Funding ID 1U19AG063744-01 
Organisation National Institutes of Health (NIH) 
Sector Public
Country United States
Start 06/2019 
End 06/2024
 
Description Artificial intelligence (AI)-driven drug discovery for rare diseases
Amount £4,800,000 (GBP)
Organisation King Abdullah University of Science and Technology (KAUST) 
Sector Academic/University
Country Saudi Arabia
Start 11/2020 
End 10/2024
 
Description Deep learning of genomic associations with Alzheimer's Disease
Amount £1,200,000 (GBP)
Organisation Janssen Pharmaceutica NV 
Sector Private
Country Belgium
Start 01/2021 
End 12/2023
 
Description Deep learning of genomic associations with Alzheimer's Disease (J&J)
Amount £400,000 (GBP)
Funding ID No reference number yet 
Organisation Johnson & Johnson 
Sector Private
Country United States
Start 05/2020 
End 05/2023
 
Description Funding for infrastructure support from Oxford Health Biomedical Research Centre - to a run an online pupil survey with participating schools in Oxfordshire (in collaboration with Oxford County Council)
Amount £35,000 (GBP)
Organisation Oxford University Hospitals NHS Foundation Trust 
Department NIHR Oxford Biomedical Research Centre
Sector Academic/University
Country United Kingdom
Start 02/2019 
End 03/2020
 
Description Research Activity in the OxWell School Survey on Mental Health and Wellbeing
Amount £25,000 (GBP)
Organisation NHS Frimley Clinical Commissioning Group (CCG) 
Sector Public
Country United Kingdom
Start 11/2021 
End 12/2022
 
Description Rising Start Initiative - stage 2
Amount £99,000 (GBP)
Funding ID No reference number 
Organisation Ono Pharmaceutical 
Sector Private
Country Japan
Start 03/2020 
End 06/2021
 
Description The Health and Well-being Pupil Survey for the COVID-19 pandemic
Amount £74,100 (GBP)
Organisation Westminster Foundation 
Sector Charity/Non Profit
Country United Kingdom
Start 04/2020 
End 10/2022
 
Description The School Health and Well-being survey: dissemination and implementation of Covid-19 pandemic findings
Amount £10,000 (GBP)
Organisation National Institute for Health Research 
Sector Public
Country United Kingdom
Start 06/2020 
End 10/2020
 
Description Unlocking evidence from electronic patient records for smart intervention of mental health disorders (EPSRC)
Amount £45,000 (GBP)
Funding ID EP/N026977/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 03/2018 
End 03/2019
 
Description Unparameterised multi-modal data, high order signatures, and the mathematics of data science
Amount £4,100,854 (GBP)
Funding ID EP/S026347/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 05/2019 
End 04/2025
 
Description Using Deep Learning to Identify and Reverse LOAD Phenotype
Amount £4,000,000 (GBP)
Organisation GlaxoSmithKline (GSK) 
Sector Private
Country Global
Start 03/2021 
End 02/2024
 
Description Virtual Brain Cloud (H2020)
Amount € 15,000,000 (EUR)
Funding ID H2020-SC1-DTH-2018-1 
Organisation European Commission 
Sector Public
Country European Union (EU)
Start 01/2019 
End 12/2022
 
Title EMIF1000 
Description We have collected 1200 blood samples of psychiatric patients and controls, and measured protein concentrations with Somalogic 
Type Of Material Database/Collection of data 
Year Produced 2019 
Provided To Others? Yes  
Impact New bio-markers of psychiatric disorders 
 
Title MAP2PD 
Description We have collected blood samples of patients with neurological disorders, and measured protein concentrations with Somalogic technology 
Type Of Material Database/Collection of data 
Year Produced 2020 
Provided To Others? Yes  
Impact We are creating bio-markers with these data samples 
 
Title OxWell School Survey Dataset 
Description OxWell is an annually-repeated cross-sectional online pupil survey. It collects a broad range of self-reported measures of mental health and wellbeing, risk and protective factors, demographics, lifestyle factors and young people's attitudes to providing sensitive data for research purposes. To date the survey has gathered 26,000 responses and is expected to collect data from a larger sample in May-July 2021. 
Type Of Material Database/Collection of data 
Year Produced 2021 
Provided To Others? Yes  
Impact We are inviting collaborators to contact us and apply to analyse the data to answer their own research questions. Currently, fully anonymous extracts of the data can be supplied to researchers following a review by the research team, and we are now seeking further funding to make the data accessible on a platform hosted by SeRP. 
URL https://www.psych.ox.ac.uk/research/schoolmentalhealth
 
Description Akrivia Health 
Organisation Akrivia Health
Country United Kingdom 
Sector Hospitals 
PI Contribution We have helped create this new University of Oxford's start-up. We have provided considerable advice and support to Akrivia during its foundation phase and beyond. Some of the leads of our MRC Pathfinder program are active members of Akrivia, occupying leading roles within the company to support it during its initial growing phase.
Collaborator Contribution Akrivia is managing computational and data resources used during our MRC Pathfinder program. They are doing so at no additional cost.
Impact Akrivia is providing very significant data management and computational support to the Natural Language Processing work of our MRC Pathfinder program. This support is ongoing, stable, and is planned to service us at no additional cost during the foreseeable future.
Start Year 2019
 
Description Alan Turing Institute 
Organisation Alan Turing Institute
Country United Kingdom 
Sector Academic/University 
PI Contribution We provide advice to Alan Turing Institute on machine learning methods applied to psychiatric disorders. We collaborate in their talk series regarding natural language processing methods in mental health reserach.
Collaborator Contribution They provide advice and further funds for us to apply machine learning methods to data collected with digital apps from patients.
Impact Further collaborations with members of the network of the Alan Turing Institute. Some of our publications presented in this report.
Start Year 2019
 
Description Berkshire West NHS Clinical Commissioning Group and the OxWell School Survey 
Organisation NHS Berkshire West CCG
Country United Kingdom 
Sector Public 
PI Contribution MRC Pathfinder iSAM team members coordinated an online pupil survey (now referred to as the OxWell School survey) and informed schools in West Berkshire, the CCG and their local authority partners of the findings, by providing brief summary reports and access to an online data summary tool.
Collaborator Contribution Berkshire West NHS CCG and their local authority partners supported the research team with the recruitment and engagement of schools in Berkshire West. A small financial contribution covers admin support with the liaison of schools, to ensure that the survey was run in accordance with the OxWell and University of Oxford ethical and data governance approvals (e.g. study instructions, information, and contracts with schools).
Impact Outputs include the OxWell School Survey dataset 2021, scientific publications, tailored school and regional reports, online summaries of results, presentations and outreach work with the public and policy partners. This collaboration involves expertise from researchers, clinicians, public health and education consultants, school head teachers and wellbeing leads, as well as collecting the views of students participating in the survey and incorporating and feedback from parents and other school staff in the research design.
Start Year 2020
 
Description Buckinghamshire Public Health and MRC Pathfinder colleagues 
Organisation Buckinghamshire County Council
Country United Kingdom 
Sector Public 
PI Contribution The research team coordinated an online pupil survey (the OxWell School Survey) and provided Buckinghamshire Public Health, their local authority partners and schools in Buckinghamshire with rapid reports on findings from the survey and access to an online data summary tool.
Collaborator Contribution Buckinghamshire Public Health supported the research team with the recruitment and engagement of schools in Buckinghamshire, and contributed to the additional costs of the survey provider in extending the survey to Buckinghamshire schools very rapidly in July 2020.
Impact This collaboration has contributed to the OxWell School survey dataset and engagement activities with the local authority, schools and pupils in Buckinghamshire.
Start Year 2020
 
Description DPUK 
Organisation MRC Dementias Platform UK
Country United Kingdom 
Sector Academic/University 
PI Contribution We have provided our blood biomarkers datasets to DPUK, in order for them to make such data available to other researchers internationally. We have further selected, managed and analysed blood samples provided by DPUK in order to create new biomarkers of psychiatric disorders.
Collaborator Contribution DPUK has provided a significant part of the blood samples used by Somalogic to measure protein concentrations. Somalogic then submitted this data back to us in order for us to build new biomarkers of psychiatric and neurological disorders.
Impact We have built several blood based biomarkers with the samples obtained through this collaboration. Several of the publications are a direct consequence of this work.
Start Year 2019
 
Description Deep Cognito 
Organisation Deep Cognito, Manchester, UK
Country United Kingdom 
Sector Private 
PI Contribution We have hired the services of Deep Cognito in order for them to develop some of the Natural Language Processing (NLP) tools used in our MRC Pathfinder program.
Collaborator Contribution Deep Cognito has developed some of the NLP tools used in this MRC Pathfinder program. They have done so at an special reduced price as an in-kind contribution.
Impact Several of our publications are the result of the NLP tools created through this collaboration. We further obtained £60,000 in funding from EPSRC's New Mind program thanks to this collaboration.
Start Year 2018
 
Description Department for Children, Adults and Health (S. Gloucestershire County Council) and Oxford MRC Pathfinder colleagues 
Organisation South Gloucestershire Council
Country United Kingdom 
Sector Public 
PI Contribution We provided expertise in making the best use of their survey data for research purposes, and also guiding policy.
Collaborator Contribution The council provided us access to their survey data to include in an analysis of our Oxfordshire data.
Impact A platform for this data is being set up on UKSeRP infrastructure.
Start Year 2019
 
Description Foster and Brown Research Ltd and MRC Oxford Pathfinder Colleagues 
Organisation Foster and Brown Research
Country United Kingdom 
Sector Private 
PI Contribution We gave them £65,000 (alongside the Oxfordshire County Council's £10,000). UPDATE: Foster and Brown have since received an additional £10,500 from us for the 2020 changes.
Collaborator Contribution They collected online pupil survey data including recruitment, technical support, data cleaning, liaison with schools, training sessions and reporting. UPDATE: Schools and partners also get access to Foster and Browns's online data summary tool (which is a big sell for the schools and LAs to take part).
Impact Outputs include - 4500 pupils' data. We're expecting another 15000 in 2020/21. Data are made accessible to participating schools and the local authority via an online data tool. UPDATE: The outputs are much higher now - the OxWell database (listed as a new output) already has 26,000 responses and we have an additional 10 local authorities on board.
Start Year 2018
 
Description Fraunhofer Society 
Organisation Fraunhofer Society
Department Fraunhofer Institute for Algorithms and Scientific Computing
Country Germany 
Sector Public 
PI Contribution We have provided blood biomarker datasets to the Franhofer SCAI institute through our joint Virtual Brain Cloud program.
Collaborator Contribution Thanks to their accumulated experience on this area, SCAI has thoroughly curated the datasets that we have provided, to a level of quality that we would have been unable to achieve.
Impact Several publications in preparation are a direct consequence on this collaboration. Also a consequence of this collaboration is significant additional funding that we have obtained from the European commission in a joint proposal between Oxford and SCAI. They have further connected us with a significant network of collaborators across European, who closely share our same research interests.
Start Year 2018
 
Description Frimley NHS Clinical Commissioning Group (East Berkshire) and Oxford Pathfinder colleagues 
Organisation NHS Frimley Clinical Commissioning Group (CCG)
Country United Kingdom 
Sector Public 
PI Contribution MRC Pathfinder iSAM team members coordinated an online pupil survey (now called the OxWell School survey) and informed schools in East Berkshire, the CCG and their local authority partners of the findings, by providing brief summary reports and access to an online data summary tool.
Collaborator Contribution East Berkshire CCG team and their local authority partners helped with the recruitment and engagement of schools in East Berkshire. Initially, a small financial contribution covered admin support with the liaison of schools, to ensure that the survey was run in accordance with the OxWell and University of Oxford ethical and data governance approvals (e.g. study instructions, information, and contracts with schools). Since November 2021, a larger financial donation supports OxWell research activity (a research assistant), including engagement work with the CCG, public health and wellbeing leads in East Berkshire.
Impact Outputs include the OxWell School Survey dataset 2021, scientific publications, tailored school and regional reports, online summaries of results, presentations and outreach work with the public and policy partners. This collaboration involves expertise from researchers, clinicians, public health and education consultants, school head teachers and wellbeing leads, as well as collecting the views of students participating in the survey and incorporating and feedback from parents and other school staff in the research design.
Start Year 2020
 
Description Funded collaboration with GSK 
Organisation GlaxoSmithKline (GSK)
Department Neuroscience (GSK)
Country United Kingdom 
Sector Private 
PI Contribution We will use neural networks to automatically extract the effect of abeta and genotypes from microscopy images of iPSC derived neurons, microglia and astrocytes. This information will then be used in high throughput drug profiling to find new drug therapies of neurological disorders.
Collaborator Contribution Funds for postdoctoral salaries and laboratory running costs.
Impact none yet
Start Year 2020
 
Description Funded collaboration with King Abdullah University 
Organisation King Abdullah University of Science and Technology (KAUST)
Country Saudi Arabia 
Sector Academic/University 
PI Contribution Using neural networks to analyse the genome and medical history of patients in UK Biobank to find new potential therapies for common neurological disorders.
Collaborator Contribution Funding to support postdoctoral salaries and laboratory running costs. Electronic health records from Saudi Arabia.
Impact None yet
Start Year 2020
 
Description Generation Scotland - University of Edinburgh 
Organisation Generation Scotland
Country United Kingdom 
Sector Charity/Non Profit 
PI Contribution We have selected and managed the blood samples provided to us by Generation Scotland. Somalogic has measured the concentration of 5000 proteins in these samples, which we have then analysed to created biomarkers of psychiatric and neurological disorders.
Collaborator Contribution The University of Edinburgh, through their Generation Scotland program, has provided a significant part of the blood samples used in our collaboration with Somalogic
Impact A good part of the samples profiled by Somalogic are coming from this collaboration with Generation Scotland. Several of the publications that resulted from our MRC Pathfinder program used these samples as their core dataset.
Start Year 2019
 
Description Generation Scotland - University of Edinburgh 
Organisation University of Edinburgh
Country United Kingdom 
Sector Academic/University 
PI Contribution We have selected and managed the blood samples provided to us by Generation Scotland. Somalogic has measured the concentration of 5000 proteins in these samples, which we have then analysed to created biomarkers of psychiatric and neurological disorders.
Collaborator Contribution The University of Edinburgh, through their Generation Scotland program, has provided a significant part of the blood samples used in our collaboration with Somalogic
Impact A good part of the samples profiled by Somalogic are coming from this collaboration with Generation Scotland. Several of the publications that resulted from our MRC Pathfinder program used these samples as their core dataset.
Start Year 2019
 
Description Gloucestershire NHS CCG and MRC Pathfinder colleagues 
Organisation NHS Gloucestershire CCG
Country United Kingdom 
Sector Public 
PI Contribution MRC Pathfinder iSAM team members coordinated an online pupil survey (now referred to as the OxWell School survey) and informed schools in Gloucestershire, the CCG and their local authority partners of the findings, by providing brief summary reports and access to an online data summary tool.
Collaborator Contribution Gloucestershire NHS CCG supported the research team with the recruitment and engagement of schools in Gloucestershire, and contributed financially (independently) to the associated costs of the survey provider.
Impact This collaboration has contributed to the OxWell School survey dataset and engagement activities with the the CCG, local authority, schools and pupils in Gloucestershire.
Start Year 2020
 
Description Johnson and Johnson 
Organisation Johnson & Johnson
Department Janssen Pharmaceuticals
Country United States 
Sector Private 
PI Contribution We are analysing the cardiovascular and immune component of psychiatric disorders with additional funds provided by Johnson and Johnson (J&J), which is a common area of scientific interest to our laboratory and to J&J.
Collaborator Contribution J&J has provided significant funds to completely support the research on this common area of interest.
Impact Several of the publications of our MRC Pathfinder program are a direct consequence of this collaboration. Future collaboration with J&J in planning stage are a direct consequence of this shared work with J&J.
Start Year 2018
 
Description MRC Pathfinder 
Organisation Dementia Pathfinders
Country United Kingdom 
Sector Charity/Non Profit 
PI Contribution I have led all the different phases of the project and closely supervised the work of my research team. I have not stared from scratch, because I have been preparing the ground for this project for years. I have completed the preliminary analyses of the studies to be included in the IPD-NMA, and have collected the IPD from a significant proportion of clinical studies. I am working with relevant stakeholders to grant me access to additional unpublished information, in collaboration with patients and clinicians and I have developed a prototype of the treatment algorithm. I have existing multidisciplinary collaborations (Salanti, Higgins, Ioannidis, Egger, Browning, De Vos) that allow me to progress on the project in timely way and I have established this collaboration with the MRC Pathfinder that will help me carry out the tasks for which I do not have direct expertise (for instance, the analysis of large observational datasets).
Collaborator Contribution MRC Pathfinder colleagues with machine learning expertise have helped design the method section of the analysis protocol and the treatment algorithm that will inform the decision making tool.
Impact Together we are developing a clinical support system (treatment algorithm), to guide clinicians, patients and carers in the shared process of decision-making in routine care. The web-based algorithm will: Utilise the predictions from the best predictive models identified during the first phase of the NIHR project. Incorporate preferences and values of clinicians and patients to identify the desired clinical outcome considering both efficacy and adverse events Generate the corresponding ranking list of personalised treatment recommendations that will inform the clinical discussion between clinicians and patients, and the final treatment decision.
Start Year 2019
 
Description Milton Keynes Council and the OxWell Survey 
Organisation Milton Keynes City Council
Country United Kingdom 
Sector Public 
PI Contribution The research team coordinated an online pupil survey (the OxWell School Survey) and provided Milton Keynes Council, their partners and schools, with rapid reports on findings from the survey and access to an online data summary tool
Collaborator Contribution Milton Keynes partners in the Public Health and Education teams, and participating schools, advised the OxWell Survey team on areas of concern in adolescent wellbeing that are relevant to measure and understand. A financial contribution covered the costs of admin support for the liaison with participating schools, to ensure that the survey was carried out in line with the university ethical and information governance guidelines (e.g. study instructions, tailored information, and contracts with schools).
Impact Outputs include the OxWell School Survey dataset 2021, scientific publications, tailored school and regional reports, online summaries of results, presentations and outreach work with the public and policy partners. This collaboration involves expertise from researchers, clinicians, public health and education consultants, school head teachers and wellbeing leads, as well as collecting the views of students participating in the survey and incorporating and feedback from parents and other school staff in the research design.
Start Year 2021
 
Description NIHR's Applied Research Collaboration (ARC) and MRC Oxford Pathfinder colleague 
Organisation National Institute for Health Research
Department NIHR CLAHRC East of England
Country United Kingdom 
Sector Public 
PI Contribution We are working together in next iteration of the survey and research and dissemination plan: Schools Mental Health Implementation Workstream, Mental Health across the Lifespan Theme, along with ARC.
Collaborator Contribution They funded the salary of one of our lead researchers and guidance.
Impact We are working together in next iteration of the survey and research and dissemination plan: Schools Mental Health Implementation Workstream, Mental Health across the Lifespan Theme, along with ARC.
Start Year 2019
 
Description New major collaboration with Johnson and Johnson 
Organisation Johnson & Johnson
Department Janssen Pharmaceutica
Country Global 
Sector Private 
PI Contribution We will be applying machine learning and neural network methods to analyse RNA sequencing from 150000 patients of UK Biobank to find new Alzheimer's disease targets
Collaborator Contribution They are funding 3 new postdoctoral researchers across the University, plus significant overhead costs. They have paid for the very substantial costs of measuring RNA sequencing on 150000 patients.
Impact Multi-disciplinary
Start Year 2020
 
Description Ono Pharma 
Organisation Ono Pharmaceutical
Country Japan 
Sector Private 
PI Contribution We have been developing new Artificial Intelligence (AI) and Neural Networks (NN) methods to analyse genomics datasets. These new algorithms have a sensitivity unmatched by any current genomics technique. Ono Pharmaceutical is then applying these algorithms to psychiatric disorders of common interest to us and Ono.
Collaborator Contribution Ono Pharma has provided significant funds to completely fund this work.
Impact Several publications in preparation are a direct consequence of this collaboration. New algorithms to analyse genomic data have been developed.
Start Year 2018
 
Description RED January & MIND exercise intervention 
Organisation Mind Association
Country United Kingdom 
Sector Charity/Non Profit 
PI Contribution This is a new collaboration between the RED January organisation, Mind, the mental health charity and the University of Oxford. We wanted to test the efficacy of the month-long 'intervention' to encourage people to exercise to benefit their mental health. A post-doc (Catherine Wheatley) adapted the before and after questionnaires in collaboration with Mind, and conducted structured interviews with >30 participants.
Collaborator Contribution Mind collects and hosts the questionnaire data which is collected from individuals who respond to RED January marketing material. We entered into a data sharing arrangement with Mind so that we can analyse the data.
Impact Data collection just completed - outputs pending
Start Year 2019
 
Description Sleepio 
Organisation Oxford Academic Health Science Network
Country United Kingdom 
Sector Charity/Non Profit 
PI Contribution Using the data provided by Sleepio, we have built Artificial Intelligence (AI) and Neural Networks (NN) algorithms able to predict future prognosis of patients with sleep problems. One of the publications recently submitted, which is part of our MRC Pathfinder program, is a direct consequence of this work.
Collaborator Contribution Sleepio has given controlled access to anonymized data of users of their digital platform. Further, they have provided advice and support to help us interpret and correctly analyse this data.
Impact One of our recently submitted publications are a direct result of this collaboration. We are further developing more accuracy AI and NN algorithms able to predict prognosis and recommend therapies with this data.
Start Year 2019
 
Description Somalogic 
Organisation SomaLogic
Country United States 
Sector Private 
PI Contribution Somalogic (https://somalogic.com/) is our main partner in our bio-markers research. They specialise on measuring protein concentration in blood samples, with a state-of-the-art platform able to obtain the concentration of 5000 proteins simultaneously. This is the service that they provide to us. The collaboration is very active and on-going.
Collaborator Contribution Our partners select and prepare the blood samples of our research participants. Somalogic then process these samples to measure protein concentration, which they then submit back to us. We use these measurements to predict patient prognosis.
Impact - Thanks to this collaboration, we have obtained the protein concentration of approximately 1200 patients and controls. We have then used these measurements to predict patient diagnosis in advance to the presentation of any behavioural symptoms. Several of the publications of our MRC Pathfinder program are a result of these samples. Future publications will still emerge from these samples - Through this collaboration we are currently processing an additional 2000 samples of psychiatric patients and controls. Once Somalogic provides to us their measurements, this dataset will be the largest proteomic cohort word-wide for some of the diseases under scrutiny. Significant publications are expected to emerge from this dataset.
Start Year 2018
 
Description UK-CRIS 
Organisation UK-CRIS, UK
Country United Kingdom 
Sector Academic/University 
PI Contribution Our Natural Language Processing (NLP) work is a significant part of the early research outputs of the UK-CRIS program. We have provided NLP tools to UK-CRIS in order for other researchers and the NHS to be able to mine the text data of the Electronic Health Records (EHRs) made accessible by UK-CRIS. We have further provided all the medical terms that we have been able to extract from these records with our NLP tools.
Collaborator Contribution UK-CRIS has made available a very significant data resource to us. This data resource consists on the EHRs of 12 Mental Health trusts across the UK, amountin to 2.6 million patients. This is to our knowledge the largest such mental health EHR internationally
Impact All our NLP publications are a direct consequence of this collaboration
Start Year 2018
 
Description University of Manchester - NLP 
Organisation University of Manchester
Country United Kingdom 
Sector Academic/University 
PI Contribution We have collaborated with Prof Goran Nenadic and his team (University of Manchester) on the Natural Language Processing (NLP) work of our MRC Pathfinder program.
Collaborator Contribution Prof Nenadic has provided advice and supervision to our NLP work.
Impact Several of the publications of our MRC program are a direct consequence of this collaboration. As a consequence of this collaboration, we have further obtained £60,000 in funding from EPSRC's New Mind program.
Start Year 2018
 
Title Protocol to test automated analysis of facial emotions as a marker of mood in patients with mood disorders 
Description Researchers in the MRC Oxford Pathfinder Mood Recognition project have developed a protocol to test automated analysis of facial emotions, using cloud-based cognitive services, combined with validated self-reported measures of mood and emotion, and mathematical models to interpret the multi-modal data. Use of the first computer-based app was approved by NHS-HRA in 2019 and has been used to collect data from patients and healthy volunteers. Following delays due to the pandemic, remote versions of the app are being refined to collect the data online following approval of an amendment to the ethical protocol. Following the end of the Pathfinder award, the project is now being funded by and EPSRC grant to Professor Terry Lyons. 
Type Diagnostic Tool - Imaging
Current Stage Of Development Early clinical assessment
Year Development Stage Completed 2019
Development Status Under active development/distribution
Impact Work related to this project and protocol has led to securing further funding from EPSRC. 
URL https://www.hra.nhs.uk/planning-and-improving-research/application-summaries/research-summaries/faci...
 
Title Emotion Recognition Apps (computer and mobile phone versions) 
Description MRC Pathfinder colleagues in the Mood Recognition project have developed one computer-interface application and one mobile phone application for multi-modal collection of emotion measures. Both versions of the app incorporate validated self-report measures of mood and emotion with brief video and audio recordings. The video can be analysed using automated emotion recognition software, such as made available as part of the Microsoft Cognitive Services, and then compared to the self-reported mood/emotions, emotion recognition task performance data, and (for patients) clinicians' ratings of mood. 
Type Of Technology Webtool/Application 
Year Produced 2019 
Impact The research team is still collecting data, which faced some delays due to the pandemic, but now developing the computer-based app to start online data collection as soon as amended ethical approval is granted. 
 
Title Machine learning genomics 
Description New machine learning methods able to improve the sensitivity with which we can analyse genomics data from psychiatric disorders. 
Type Of Technology New/Improved Technique/Technology 
Year Produced 2019 
Impact MSc students projects and publications 
 
Title New Mind NLP library 
Description In collaboration with the University of Manchester and Deep Cognito (Manchester), we have developed a Java library to extract medical information from text as recorded in Electronic Health Records. 
Type Of Technology New/Improved Technique/Technology 
Year Produced 2019 
Impact Some of our publications are a result of this software. Our research team and collaborators in the UK-CRIS network are using this software. 
 
Title True Colours platform 
Description True Colours is a web application that allows patients and participants to self monitor their mood and associated metrics (e.g. sleep, life events, activity etc). It has been in use for >10 years and the Pathfinder grant has enabled its continued development and maintenance. 
Type Of Technology Webtool/Application 
Year Produced 2019 
Impact True Colours has been used by >36000 patients/participants as part of research studies, trials and clinical service. There have been >50 publications arising from its use and a paper was published this year describing the impact (Goodday et al, 2019 - see publication list). 
URL https://oxfordhealth.truecolours.nhs.uk/www/en/
 
Company Name Akrivia Health (Cristal Health) 
Description With over 440 billion datapoints, agreements with 12 NHS trusts, and more than 3 million patient records, Akrivia has the largest and richest mental health dataset to accelerate research for better treatments and care. 
Year Established 2019 
Impact The company is making Electronic Health Records (EHRs) accessible for research purposes. A thorough data access process has been established for this aim, with close supervision from patients and NHS trusts.
Website https://akriviahealth.com/
 
Description : Feedback of findings of Oxfordshire Online Pupil Survey to school mental health leads as part of Oxford Health NHSFT outreach event to Trailblazer schools. 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Schools
Results and Impact As described above
Year(s) Of Engagement Activity 2020
 
Description CAMHS evaluation Dissemination event 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Professional Practitioners
Results and Impact Dissemination activity informing attendees of results of evaluation of CAMHS including some key data from the survey that was conducted as partof pathfinder
Year(s) Of Engagement Activity 2019
 
Description Chair at NICE meeting to discuss guidelines for depression 
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 Chaired a meeting to discuss the NICE guidelines for treating depression.
Year(s) Of Engagement Activity 2019
 
Description Chaired a session on 'MRC Pathfinder Session 2' at MQ Mental Health Conference in Edinburgh, Scotland (September 2019) 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Postgraduate students
Results and Impact John chaired a session on the Edinburgh, KCL and other Pathfinder award presentations at the MQ Mental Health Conference in Edinburgh, Scotland (September 2019). He took the place of Prof Andrew McIntosh at very short notice. The purpose was to increase awareness of one's own work and network with other Pathfinder award winners. The impact was gaining new insight into other Pathfinder projects' and new strategies for developing a platform.
Year(s) Of Engagement Activity 2019
URL https://www.mqmentalhealth.org/articles/data-science-meeting-2019-agenda
 
Description Mental Health TRC Medical Informatics Workstream 
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 Mental Health TRC Medical Informatics workstream committee member. A 'Medical Informatics' MH-TRC workstream was proposed in early 2020 and leads (Ann John, John Ainsworth) appointed in September 2020. 'Medical Informatics' for the purposes of the MH-TRC workstream is defined as routine information collected electronically in health services. While mental health service data are of particular interest, a volume of mental healthcare occurs in primary care and emergency departments, so information from these sources is equally relevant.

Our aim is to facilitate the mental health research community to make effective use of routinely collected data and digital technologies.
Year(s) Of Engagement Activity 2020
URL https://www.nihr.ac.uk/partners-and-industry/industry/collaborate-with-us/mental-health-trc.htm
 
Description Presentation at the Oxford Child and Adolescent Psychiatry Seminar series to policy makers and other audiences 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Policymakers/politicians
Results and Impact Presentation to more than 100 policy makers, practitioners and researchers on some of the findings from the OxWell School Survey from data collected during the first UK partial schools closures in June-July 2020. Many questions and comments at the end of the presentation came from public health and clinical commissioners, facilitating alignment of our analyses with public health priorities.
Year(s) Of Engagement Activity 2020
URL https://www.youtube.com/watch?v=c-h7-cCPOhk
 
Description Presentation of results highlights from the 2020 OxWell School Survey to Gloucestershire CCG and Council 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Policymakers/politicians
Results and Impact Presentation of results highlights from the OxWell School Survey to commissioners and partners working with Gloucestershire Council and Gloucestershire NHS CCG. The presentation focused on self-reported wellbeing and experience of lockdown from pupils in Gloucestershire schools, with some comparisons to the broader regional OxWell 2020 data.
Year(s) Of Engagement Activity 2020
 
Description Presentation of results highlights from the OxWell School Survey to Berkshire CCGs and local authorities 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Policymakers/politicians
Results and Impact Presentation of results highlights from the OxWell School Survey to commissioners and partners working with Berkshire East and West Council and CCGs. The presentation focused on self-reported wellbeing and experience of lockdown from pupils in Berkshire schools, with some comparisons to the broader regional OxWell 2020 data.
Year(s) Of Engagement Activity 2020
 
Description Presentation of results highlights from the OxWell School Survey to Buckinghamshire Council 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Policymakers/politicians
Results and Impact Presentation of results highlights from the OxWell School Survey to commissioners and partners working with Buckinghamshire Council. The presentation focused on self-reported wellbeing and experience of lockdown from pupils in Buckinghamshire schools, with some comparisons to the broader regional OxWell 2020 data.
Year(s) Of Engagement Activity 2020
 
Description Presentation of survey results. Youth in Mind conference. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Third sector organisations
Results and Impact Presentation of research plan with keynote presentation to delegates from children's services across all sectors in Oxfordshire
Year(s) Of Engagement Activity 2019
URL https://www.oxfordshiremind.org.uk/youth-in-mind-conference/
 
Description Presentation to schools and parents at "Young Resilient Minds" event 
Form Of Engagement Activity Participation in an activity, workshop or similar
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Schools
Results and Impact Presentation of preliminary pupil survey results to local schools, service providers, third sector and parents at "Resilient Young Minds" event organised at Bladon Primary School.
After the presentation, I (Karen Mansfield) gave a short interview to local news.
Year(s) Of Engagement Activity 2019
URL https://www.youtube.com/watch?v=vK_cFOky22c
 
Description Public seminar for Department of Education 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Postgraduate students
Results and Impact This is made into a podcast and presentation of findings from survey has been included
Year(s) Of Engagement Activity 2020
 
Description School governor meeting 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Schools
Results and Impact Talk to school governors including results of pathfinder survey
Year(s) Of Engagement Activity 2019
 
Description School meeting with governors and pastoral team leads 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach Local
Primary Audience Schools
Results and Impact In depth examination of survey findings of relevance to the school
Year(s) Of Engagement Activity 2019
 
Description Supported Karen Mansfield (Postdoc) Presentation on Pathfinder at MhDEL workshop (Mental Health Data Ethics Leadership - Ethics and Governance in Data Linkage Workshop) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Other audiences
Results and Impact Karen Mansfield (Postdoc) presented her talk on opportunities and challenges for linking adolescent mental health data in the UK - the purpose was to get feedback from others working in the area and beyond to understand the issues from multiple perspectives, including HRA and funders. Outcome - we got useful feedback and contacts to move this work forward in collaboration with other researchers across the UK.
Year(s) Of Engagement Activity 2019
 
Description Thames Valley Strategic Health meeting 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Professional Practitioners
Results and Impact Discussion with key practitioner with regards to the long term plans for childrens services
Year(s) Of Engagement Activity 2020
 
Description child Psychiatry Regional Meeting 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Professional Practitioners
Results and Impact Presentation of pathfinder results to Child Psychiatrists
Year(s) Of Engagement Activity 2019