AIDAVA: Difference between revisions

From Research Wiki
← Older edit
Undo revision 129 by Admin 3julmthh (talk)
Tag: Undo
Restored content after accidental blanking
 
Line 1: Line 1:
__NOTOC__
__NOTOC__
AI-powered Data Curation & Publishing Virtual Assistant. EU Horizon Europe research project automating curation and publishing of personal health data using AI.
AI-powered Data Curation & Publishing Virtual Assistant. EU Horizon Europe research project automating curation and publishing of personal health data using AI.


== Overview ==
AIDAVA prototypes and tests an AI-powered virtual assistant that maximizes automation of data curation and publishing of unstructured and structured, heterogeneous health data. The assistant includes a backend library of AI-based data curation tools and a frontend based on human-AI interaction modules.<ref>European Commission CORDIS, Grant 101057062: "AI powered Data Curation & Publishing Virtual Assistant" — https://cordis.europa.eu/project/id/101057062</ref>
AIDAVA prototypes and tests an AI-powered virtual assistant that maximizes automation of data curation and publishing of unstructured and structured, heterogeneous health data. The assistant includes a backend library of AI-based data curation tools and a frontend based on human-AI interaction modules.<ref>European Commission CORDIS, Grant 101057062: "AI powered Data Curation & Publishing Virtual Assistant" — https://cordis.europa.eu/project/id/101057062</ref>


== Key Facts ==
== Key Facts ==
{|
{| class="wikitable"
|-
|-
! Detail !! Value
! Detail !! Value
Line 13: Line 11:
| Grant ID || [https://cordis.europa.eu/project/id/101057062 101057062]
| Grant ID || [https://cordis.europa.eu/project/id/101057062 101057062]
|-
|-
| Funding Call || HORIZON-HLTH-2021-TOOL-06
| Funding Call || HORIZON-HLTH-2021-TOOL-06-03
|-
| Total Cost || €7,720,620
|-
|-
| EU Contribution || €7,720,620 (100% funded)
| Total Cost || EUR 7,720,615
|-
|-
| Start Date || September 1, 2022
| Start Date || September 1, 2022
Line 23: Line 19:
| End Date || August 31, 2026
| End Date || August 31, 2026
|-
|-
| Duration || 4 years
| Duration || 48 months
|-
|-
| Partners || 13 from 9 countries
| Partners || 13-14 from 9 countries
|-
|-
| Type || Horizon Europe Research and Innovation Action (RIA)
| Coordinator || [[Maastricht University]], Remzi Celebi — remzi.celebi@maastrichtuniversity.nl
|-
|-
| Website || [https://aidava.eu aidava.eu]
| Website || [https://aidava.eu aidava.eu]
|-
| Documents || [[AIDAVA Documents]] — all PDFs with summaries
|-
| Contacts || [[AIDAVA Contacts]] — partner people and emails
|}
|}


== Vision ==
"Curate once, reuse many times" — supporting patients, clinical care providers and clinical researchers from the same curated data.<ref>AIDAVA official website — https://aidava.eu</ref>
"Curate once, reuse many times" — supporting patients, clinical care providers and clinical researchers from the same curated data.<ref>AIDAVA official website — https://aidava.eu</ref>


The project addresses the core problem that integrated, high-quality personal health data represents a potential wealth of knowledge for healthcare systems, but there is no reliable conduit for this data to become interoperable, AI-ready and reuse-ready at scale across institutions, at national and EU level.<ref>European Commission CORDIS, Grant 101057062, project summary — https://cordis.europa.eu/project/id/101057062</ref>
The project addresses the core problem that integrated, high-quality personal health data represents a potential wealth of knowledge for healthcare systems, but there is no reliable conduit for this data to become interoperable, AI-ready and reuse-ready at scale.<ref>European Commission CORDIS, Grant 101057062, project summary — https://cordis.europa.eu/project/id/101057062</ref>


== Technology Pillars ==
== Technology Pillars ==
Line 44: Line 43:


== Use Cases ==
== Use Cases ==
# '''Breast cancer patient registries''' — structured registry data curation
# '''Breast cancer patient registries''' — structured registry data curation across 3 university hospitals
# '''Longitudinal health records for cardiovascular patients''' — integrating heterogeneous data sources over time
# '''Longitudinal health records for cardiovascular patients''' — integrating heterogeneous data sources over time


Both tested in three languages with hospitals and emerging personal data intermediaries.<ref>European Commission CORDIS, Grant 101057062, project summary — https://cordis.europa.eu/project/id/101057062</ref>
Both tested in three languages (Dutch, German, Estonian) with hospitals and emerging personal data intermediaries.<ref>European Commission CORDIS, Grant 101057062 — https://cordis.europa.eu/project/id/101057062</ref>


== Solution Architecture ==
== Evaluation Results (from March 2025 Flash) ==
* '''Data cleaning machine''' — orchestrating multiple AI-based tools to automate curation
* 83 patients recruited; 70 completed evaluation
* '''Personal Health Knowledge Graph (PHKG)''' — universal semantic representation of all personal health data
* G1 prototype (June 2024) tested in Estonia, Austria, Netherlands
* '''Conversational AI assistant''' — engages patients, with explainability capabilities
* 45% of documents curated automatically
* '''Metadata capture''' — on data sources to support automation within a formalised Data Transfer Specification
* Average 20 minutes per document (too slow)
* Tools orchestrated include: OCR, syntactic transformation, semantic transformation, entity deduplication, NLP, feature extraction from imaging<ref>AIDAVA, "Vision and Key Facts" — https://aidava.eu/about/vision-and-key-facts</ref>
* Usability good, but explanations suboptimal and benefits unclear
* G2 delivery: end 2025, testing early 2026
* Project end: August 2026


== Impact ==
Full document summaries: [[AIDAVA Documents]]
* Decrease workload of clinical data stewards through increased automation
* Improve effectiveness of clinical care through high-quality data
* Support clinical research with reusable, interoperable data
* Long-term: democratise participation in data curation by citizens/patients
* Support delivery of the European Health Data Space (EHDS)<ref>European Commission CORDIS, Grant 101057062, project summary — https://cordis.europa.eu/project/id/101057062</ref>


== Partners ==
== Partners ==
Line 76: Line 72:
| [[Egnosis]] || Romania || Health data intermediary
| [[Egnosis]] || Romania || Health data intermediary
|-
|-
| [[Ontotext]] || Bulgaria || Knowledge graph technology
| [[Ontotext]] (Sirma AI EAD) || Bulgaria || Knowledge graph technology
|-
|-
| [[Averbis]] || Germany || NLP and text mining
| [[Averbis]] || Germany || NLP and text mining
Line 84: Line 80:
| [[North Estonia Regional Hospital]] || Estonia || Clinical partner (use case)
| [[North Estonia Regional Hospital]] || Estonia || Clinical partner (use case)
|-
|-
| [[European Cancer Patient Coalition]] || Belgium || Patient advocacy
| [[European Cancer Patient Coalition]] || Belgium || Patient advocacy (until May 2024)
|-
|-
| [[European Heart Network]] || Belgium || Patient advocacy (cardiovascular)
| [[European Heart Network]] || Belgium || Patient advocacy (cardiovascular)
Line 90: Line 86:
| [[B!LOBA]] || Belgium || Data management
| [[B!LOBA]] || Belgium || Data management
|-
|-
| [[DFP Research]] || Spain || Research partner
| [[DFP Research]] / Data for Patients || Spain || Research partner (since Oct 2024)
|-
|-
| [[EURICE]] || Germany || Project management
| [[EURICE]] || Germany || Project management
|-
| [[MIDATA Cooperative]] || Switzerland || Data cooperative
|-
| [[RISE]] Research & Innovation Services || Croatia || Innovation services
|}
|}


All partners identified via OpenAIRE/CORDIS project registry.<ref>OpenAIRE, AIDAVA project 101057062 — https://api.openaire.eu/search/projects?keywords=AIDAVA</ref>
All contacts: [[AIDAVA Contacts]]


== Related Topics ==
== Related ==
* [[Knowledge Graphs in Health]]
* [[Knowledge Graphs in Health]]
* [[Interoperability]]
* [[Interoperability]]
* [[Longitudinal Journals]]
* [[Longitudinal Journals]]
* [[EU Regulation]]
* [[AIDAVA Documents]]
* [[AIDAVA Contacts]]


== External Links ==
== External Links ==
* [https://aidava.eu Official AIDAVA website]
* [https://aidava.eu Official website]
* [https://cordis.europa.eu/project/id/101057062 CORDIS project page]
* [https://cordis.europa.eu/project/id/101057062 CORDIS]
* [https://doi.org/10.3030/101057062 DOI: 10.3030/101057062]
* [https://doi.org/10.3030/101057062 DOI]


<references />
<references />
[[Category:Project]]
[[Category:Project]]
[[Category:EU]]
[[Category:EU]]

Latest revision as of 19:02, 13 April 2026

AI-powered Data Curation & Publishing Virtual Assistant. EU Horizon Europe research project automating curation and publishing of personal health data using AI.

AIDAVA prototypes and tests an AI-powered virtual assistant that maximizes automation of data curation and publishing of unstructured and structured, heterogeneous health data. The assistant includes a backend library of AI-based data curation tools and a frontend based on human-AI interaction modules.[1]

Key Facts[edit]

Detail Value
Grant ID 101057062
Funding Call HORIZON-HLTH-2021-TOOL-06-03
Total Cost EUR 7,720,615
Start Date September 1, 2022
End Date August 31, 2026
Duration 48 months
Partners 13-14 from 9 countries
Coordinator Maastricht University, Remzi Celebi — remzi.celebi@maastrichtuniversity.nl
Website aidava.eu
Documents AIDAVA Documents — all PDFs with summaries
Contacts AIDAVA Contacts — partner people and emails

"Curate once, reuse many times" — supporting patients, clinical care providers and clinical researchers from the same curated data.[2]

The project addresses the core problem that integrated, high-quality personal health data represents a potential wealth of knowledge for healthcare systems, but there is no reliable conduit for this data to become interoperable, AI-ready and reuse-ready at scale.[3]

Technology Pillars[edit]

  1. Automation of quality enhancement and FAIRification of collected health data, in compliance with EU data privacy
  2. Knowledge graphs with ontology-based standards as universal representation — each Personal Health Knowledge Graph (PHKG) is an instance of a common reference knowledge graph based on ontologies derived from SNOMED, HL7 FHIR resource profiles, LOINC, and other domain-specific terminologies[4]
  3. Deep learning for information extraction from narrative content — NLP developed in three languages
  4. AI-generated explanations during the process to increase users' confidence (explainability)[5]

Use Cases[edit]

  1. Breast cancer patient registries — structured registry data curation across 3 university hospitals
  2. Longitudinal health records for cardiovascular patients — integrating heterogeneous data sources over time

Both tested in three languages (Dutch, German, Estonian) with hospitals and emerging personal data intermediaries.[6]

Evaluation Results (from March 2025 Flash)[edit]

  • 83 patients recruited; 70 completed evaluation
  • G1 prototype (June 2024) tested in Estonia, Austria, Netherlands
  • 45% of documents curated automatically
  • Average 20 minutes per document (too slow)
  • Usability good, but explanations suboptimal and benefits unclear
  • G2 delivery: end 2025, testing early 2026
  • Project end: August 2026

Full document summaries: AIDAVA Documents

Partners[edit]

Organization Country Role
Maastricht University Netherlands Coordinator
KU Leuven Belgium Research partner
i-HD Belgium Health data standards and quality
Egnosis Romania Health data intermediary
Ontotext (Sirma AI EAD) Bulgaria Knowledge graph technology
Averbis Germany NLP and text mining
Medical University of Graz Austria Clinical partner
North Estonia Regional Hospital Estonia Clinical partner (use case)
European Cancer Patient Coalition Belgium Patient advocacy (until May 2024)
European Heart Network Belgium Patient advocacy (cardiovascular)
B!LOBA Belgium Data management
DFP Research / Data for Patients Spain Research partner (since Oct 2024)
EURICE Germany Project management
MIDATA Cooperative Switzerland Data cooperative
RISE Research & Innovation Services Croatia Innovation services

All contacts: AIDAVA Contacts

Related[edit]

External Links[edit]

  1. ↑ European Commission CORDIS, Grant 101057062: "AI powered Data Curation & Publishing Virtual Assistant" — https://cordis.europa.eu/project/id/101057062
  2. ↑ AIDAVA official website — https://aidava.eu
  3. ↑ European Commission CORDIS, Grant 101057062, project summary — https://cordis.europa.eu/project/id/101057062
  4. ↑ AIDAVA, "Vision and Key Facts" — https://aidava.eu/about/vision-and-key-facts
  5. ↑ AIDAVA, "Solution" — https://aidava.eu/about/solution
  6. ↑ European Commission CORDIS, Grant 101057062 — https://cordis.europa.eu/project/id/101057062
Retrieved from "https://wiki.stamen.health/index.php?title=AIDAVA&oldid=142"