MyDiabetesIQ – phase 0

• Research type

  Research Study

• Full title

  MyDiabetesIQ: the use of artificial intelligence to enhance decision support for diabetes - Phase O: Data extraction, pseudonymisation and analysis

• IRAS ID

  268895

• Contact name

  Nicholas Conway

• Contact email

  n.z.conway@dundee.ac.uk

• Sponsor organisation

  My Way Digital Health

• Duration of Study in the UK

  1 years, 8 months, 30 days

• Research summary

  Diabetes affects approximately 8% of the UK population and accounts for 10% of current NHS spending. Diabetes care in Scotland relies on a series of managed clinical networks supported by a national informatics platform, SCI-Diabetes.
  
  Both clinicians and patients can access SCI-Diabetes: clinicians access it directly and can interrogate patient data; patients can access the interactive website MyDiabetesMyWay (MDMW), developed within NHS Scotland, that acts as a patient portal, allowing users to access their own clinical data. However, data functionality is limited and both clinicians and patients desire improvements to inform personalised treatment decisions and tailored self-care advice to support the prevention of complications.

  Using pseudonymised data extracted from NHS Tayside’s SCI-Diabetes (30,000 individuals) and MDMW users (3,000 individuals), this study aims to develop new stratification and prediction models by using multi-task machine learning. Machine learning is necessary due to the nature and scope of the dataset and will identify predictors of diabetes-related clinical outcomes and complications e.g. glycaemic control, peripheral vascular disease, cardiovascular risk factors etc.
  
  The models developed in this study can then be used to support decision-making for healthcare professionals and patients (the models will be incorporated into new clinician- and patient-facing dashboards which will be developed during later research phases, outside the scope of this application).

  By using this approach this study aims to significantly improve health outcomes, reduce side-effects, enable early intervention reducing development of complications and associated morbidity and mortality. As a result, this will significantly improve quality of life and reduce the cost of treatment of diabetes in Scotland.

• REC name

  South East Scotland REC 01

• REC reference

  19/SS/0113

• Date of REC Opinion

  31 Oct 2019

• REC opinion

  Further Information Favourable Opinion