Transform the response to
Hematologic
disease by seizing the power of AI
THE CHALLENGE
Today, Haematological diseases (HDs) represent an economic burden for European society of around €22.5 billion per year, according to the European Hematology Association (EHA).
The overall number of patients affected with a haematological disease is important for instance, haematological malignancies account for about 5% of cancers. In addition, most HDs can cause chronic health problems and many of them are life-threatening conditions. The majority of these hematologic disorders, either inherited or acquired, have a genetic background.
GENOMED4ALL intends to impact the diagnosis, treatment definition and outcomes prediction processes through the development of beyond-status-quo AI tools powered by the pooling of genomic, clinical data and other “-omics” health data. The objective is to deliver value across the whole value chain of the healthcare sector, achieving higher quality treatments for patients, increasing efficiency with faster diagnosis, all at a lower cost.
STRATEGY AND SOLUTION
GENOMED4ALL project will rely on datasets of both private origin and publicly available. Initially by leveraging ERN-EuroBloodNet healthcare information and repositories, and then incorporating other repositories from public or private sources. The exploited data will be both retrospective and prospective ones, in order to enrich the value of the trained AI tools.
These datasets will be standardized internally according to various established standards (SNOMED-CT, ICD10, HL7 FHIR, VCF, Phenopackets) and will be pooled in the developed platform in a GDPR compliant framework. As the envisioned technologies need to be tested on different operational environments and cover diverse user needs, three distinct clinical use cases have been identified:
– Oncological disease: Myelodysplastic syndromes (MDS)
– Oncological disease: Multiple Myeloma (MM)
– Non-Oncological disease: Sickle Cell Disease (SCD)
Project Name
Genomics and Personalized Medicine for all though Artificial Intelligence in Haematological Diseases
Funding scheme
HORIZON 2020 RIA – Research and Innovation action Grant agreement ID: 101017549
Topic
DT-TDS-04-2020 – AI for Genomics and Personalised Medicine
Budget
€ 9 999 063, 75
Duration
January 2021 – December 2024
Project website: genomed4all.eu
Numbers
86+
clinical repositories with genomics data connected across 15 eu countries
15%
ACCURACY IMPROVEMENT IN SPECIFIC GENOMIC MARKETS FOR PROGNOSIS AND TREATMENT
80%
inCREASE IN THE ADOPTION OF open standards for -omics data per clinical site
20%
Time reduction in AI ANALYSIS AND MODEL TRAINING THROUGH SUPERCOMPUTING
DATAWIZARD ACTIVITIES
Our Involvement
Quality & Risk Management
Quality assurance and risk management by prevention/mitigation of risks, throughout the project’s duration, monitoring project variables as cost, time and effort, measuring progress against project plans and carry out and manage changes.
Design
Design of user, technical and clinical requirements: Implement a human-centred, design thinking and collaborative, co-creation approach to designing project clinical and technological outcomes.
Technical Requirements
Assessment of all data repositories and additional data sets identified for the use in GENOMED4ALL use cases. Development of a strategy to build a GDPR-compliant data infrastructure. Provide suggesting guidelines for standardization of genomic data.
Exploitation
Development and undertaking of an effective strategy to exploit the GENOMED4ALL’s innovations for research, standardization and commercial purposes. Engagement with multiple stakeholders of the EU ecosystem to stimulate the adoption of the project’s knowledge, at all levels.