HEALTHY AGEING, stage project, multi-morbidity

WORK PACKAGE 6 OBJECTIVE

With a robust public and patient-focused process, WP6 will co-create an AI-supported prevention and care approach. This will involve developing new FAIR and transparent AI models, which are generalisable for use for across the EU, and estimating model uncertainty to enable AI-guided clinical decisions.

WHY IS WP6
IMPORTANT?

AI has real potential to offer targeted, data-driven support to clinicians and patients in making healthcare decisions. By pulling from public health data, refined in projects like STAGE, we can spot patterns and trends in someone’s long-term health.

STAGE aims to innovate in this field whilst also making sure the technology is deeply understood, fully transparent, and goes forward with the support of citizens and healthcare professionals around the EU.

healthy ageing, stage project

WP6 Activities and Interlinkages

WP6 is closely aligned with WP7 in developing technology interventions that can benefit patients and clinicians in the EU. It will also draw from other work packages in establishing its parameters and building an accurate, informed AI model. For example, it will draw from WP2, WP3, WP4 and WP5 when refining it’s AI models. All work packages will be involved in ensuring the model is fair for all (T6.3). WP6 tasks include:

T6.1

Requirements Gathering

Continually examining what is required when creating a trustworthy AI-assisted tool to predict and prevent ageing with multi-morbidity. Co-creation groups will focus on designing transparent, fair, robust models. Data will be drawn from the work of WP2, WP3, WP4 and WP5

T6.2

Baseline AI Models

Building a set of baseline models for different types of ageing and alternative responses to interventions. Different methods of modelling with be compared and contrasted, and input predictors – factors like the environment – will be worked in.

T6.3

Fairness

Testing baseline models for fairness (as identified in T6.1) with full bias mitigation measures. These models will be optimised to increase fairness across groups and countries, using data from different populations and health centres. 

T6.4

Robustness and Generalisability

Assessing baseline models, in line with feedback from T6.1, on their heterogeneity and analysing for robustness and generalisability across datasets and countries, with continual assessment and calibration for different European regions.

T6.5

Explainable Interfaces

Exploring several ‘explainability’ approaches with potential users and clinical, digital health and ethics experts, with feedback being used to tailor select approaches to AI explainability, with clear visual interfaces.

T6.6

Uncertainty Estimation

Providing uncertainty / confidence scores for every prediction made by the AI tools and estimating the impact of missing any key data (such as social indicators) on predictions. Accuracy results will be displayed within the user interface.

WP6 KEY OUTPUTS

Consultation and Co-Creation

Status: Ongoing

Generalisability of AI

Status: To be developed

Explainable AI Interfaces

Status: To be developed

New AI Models

Status: Ongoing

Fairness of AI

Status: To be developed

Uncertainty Estimates

Status: To be developed

WP6 STAKEHOLDER
IMPACTS

  • A thorough consultation on AI models will allow stakeholders to directly shape the AI tools produced by STAGE, ensuring they are built from the ground-up with both patients and clinicians in mind.
  • The creation of usable, transparent AI models will help spread the impact of STAGE, by making this tool available for clinicians and researchers in other countries, encouraged by a ‘generalised’, EU-wide methodology.
  • The development of AI models will help researchers in other projects develop their own, with the principals of fairness and transparency at the heart of the creation process. This will enable our stakeholders to directly benefit from new technology, whilst ensuring that further development of AI is built on solid ethical and practical principles.
healthy ageing, stage project

WP6 PARTNERS

With support from many of the STAGE partners, WP6 tasks are being undertaken by:

University of Barcelona

Work Package Lead

Spain

SRDC

Work Package Partner

Turkey

Ab.Acus

Work Package Partner

Italy

University of Oulu

Work Package Partner

Finland

“AI tools and modelling could help offer clinicians targeted, data-driven guidance to help inform tailored health interventions across Europe. As with any new technology,
the benefits must be balanced with a transparent, fair, accessible philosophy to ensure the advantages are felt most widely.”

Marina Camacho, WP6 Lead

Meet some of the WP6 partners in the video interview below and learn more about the work conducted by the work package:

Previous:

WP5

Biological hallmarks of ageing

Next:

WP7

Person-centred applications and solutions for clinicians and ageing citizens

SUBSCRIBE TO OUR NEWSLETTER

Join our mailing list to receive the latest project news

Privacy

Your subscription

By subscribing to this list, you accept to receive information about events, news and/or newsletters about the STAGE project. The personal data collected will only be used to send you the newsletter(s) you chose to opt in to above. You can change your mind at any time by clicking the "unsubscribe" link in the footer of any email you receive from us. We will treat your information with respect.

Powered by  GDPR Cookie Compliance
Privacy Overview

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.