EGI Federation Home
Health & Medicine

LETHE

Rethinking Dementia: How Technology Helps Change the Course of Cognitive Decline

About

As the world’s population ages, the number of people living with dementia grows. Dementia has long been considered to be neither preventable nor treatable; however, while the underlying illnesses are not curable, today we know that the disease course might be modifiable with good preventive interventions at an early time point.

LETHE established novel digital biomarkers for early detection of risk factors, based on unobtrusive ICT-based passive and active monitoring. It implemented a digital-enabled intervention for cognitive decline prevention based on the evolution of a successful protocol (FINGER study) evolving into an ICT-based preventive lifestyle intervention through individualised profiling, personalised recommendations, feedback and support – FINGER 2.0 – well targeted on a population stratified by cost-effective biological biomarkers.

LETHE produced a more personalised risk factor prevention programme for people in the early stages of cognitive decline, enabling them to lead active and healthy lifestyles. The expansion of digital technology-based health prevention approaches, by reaching large populations, can save healthcare systems money on costly traditional interventions and bring benefits to society as a whole.

The Challenge

Dementia affects 55 million individuals worldwide, with nearly 10 million new cases every year, making it the seventh leading cause of death among older people globally. LETHE’s challenge is to implement a personalised prediction and intervention model for the early detection and reduction of risk factors that cause dementia, based on AI and distributed Machine Learning. A critical factor is that the target subjects are older people, who are not familiar with technology and may have problems accessing and understanding it.

LETHE implemented a data-driven risk factor prediction model for older individuals at risk of cognitive decline, novel digital biomarkers and a digital-enabled intervention based on the evolution of the FINGER study​. This has been validated with a study of over 160 participants. LETHE uses different interaction technologies to track data and provide participants with personalised suggestions for brain health, which is a challenge for the elderly population that is not accustomed to this technology.

The results of the project have created meaningful current and future impact:

  • Potential to modify the course of disease with early preventive actions.​
  • Benefits of information for persons at risk and care professionals regarding LETHE prediction, prevention and intervention.
  • Bring long-term positive effects on welfare and socio-economic systems of the EU. Save costs on the health system (by disease prevention and interventions on early risk factors related to chronic diseases) as well as increase the Quality of Life (QoL) of the older population and Quality of Life in Alzheimer’s Disease (QoL-AD) for people with dementia.
  • Collection and harmonisation of active and passive observational data for new data-driven prevention models.

EGI and LETHE: The Solution

The project was divided into two phases, which follow different approaches:

A retrospective knowledge base used to generate an initial prediction model based on four different data sets provided by the clinical partners of the LETHE consortium.

Prospective data collected from variable apps, tools and wearables. Data from these different sources has to be harmonised for the AI Modules. These create reports, visualisations and information that clinicians use for risk factor identification.

The LETHE project uses the EGI FedCloud and other EOSC services as an infrastructure solution. The EGI FedCloud provides both a comprehensive sensitive data management solution for data processing and computing services, and a platform to run bespoke project applications.

EGI services are used as follows:

  • Cloud Compute is responsible for providing the hardware necessary to run the different components. This is done in the form of Virtual Machines and disks, from a virtual allocation of vCPUs, RAM, networking and storage.
  • Infrastructure Manager supports the deployment of the platform by creating specific Virtual Machines, configuring the Kubernetes clusters and installing software. IM enables the paradigm of Infrastructure-as-Code (IaC), an approach that allows the management of the infrastructure hardware resources with source code. For this purpose, the necessary virtual infrastructure is defined in templates or text files that serve as software artefacts. IM has been instrumental in enabling the platform to be deployed in a matter of minutes and with minimal human intervention.
  • Online Storage provides the means to create large encrypted volumes necessary to store the data securely, and available for the components of the LETHE infrastructure.
  • Check-in implements the authentication and authorisation required in the project. This involves the authentication of user participants from their respective smartphone devices, the authentication of clinicians and the intercommunication of the different components of the LETHE platform. Check-in also helps organise the community and categorise users according to their role in the project.
  • DataHub serves as the data management platform for results and as the repository for the retrospective data of Phase I.

Services Provided by EGI

Run virtual machines on-demand with complete control over computing resources

Use cloud orchestrator to deploy and configure complex virtual infrastructures

Store, share and access your files and their metadata on a global scale

Login with your own credentials

Access key scientific datasets in a scalable way

The successful completion of the LETHE project demonstrates how European digital infrastructure can drive innovation in dementia prevention. EGI services were instrumental in orchestrating our entire backend - from secure authentication via EGI Check-In to hosting our AI prediction services on EGI FedCloud infrastructure. This robust foundation enabled us to achieve a 97.4% participant retention rate in our 24-month clinical trial with 156 participants, while maintaining the highest standards of data security and compliance throughout.

Sten Hanke, LETHE Project Coordinator

Key Numbers

122 vCPUs
488 GB RAM
16 TB Storage
2 Kubernetes clusters
500 GB

DataHub Space

~200 users

on EGI Check-in and 16 groups

The Providers

The following providers from the EGI Federation contributed to LETHE’s success:

  • IN2P3-IRES (CNRS) – Providing Cloud Compute, DataHub
  • GRNET – Providing Check-in
  • UPV – Providing Infrastructure Manager

Supporting Projects

Related News

Loading