Initiative file / Health
LIDIA: predictive models in EDUS
Evidence record v2
10 / 31
- Institution
- CCSS
- Editorial classification
- Verified adoption
- Initiative type
- AI component
- Documented phase
- Pilot
- Relationship to AI in the source
- Explicit relationship to AI
- Execution evidence
- Confirmed
- First evidence
- Last verified
- Next review: Nov 19, 2026
Documented scope
What it is
Internal program of four predictive health models (type 2 diabetes, lung health, acute coronary syndrome and breast cancer) running on EDUS. Identifies at-risk populations so clinics can reach out before formal diagnosis.
Objective stated by the institution
Identify people at elevated disease risk through predictive models using EDUS information.
Known timeline
Only milestones with an explicit date in the sources or editorial review are shown.
- First evidence
- Pilot start
- Latest editorial verification
- Next scheduled review
Evidence reading
Each dimension is assessed separately. Undetermined marks an evidence gap; not applicable means the dimension does not correspond to the documented object.
What is confirmed
Confirmed evidence is detailed in the matrix and documented results; there are no additional notes in this field.
What could not be determined
- No independently audited public metrics for accuracy, sensitivity or false positives were found.
Documented results
Results are published only when a source traceably attributes them to this initiative.
Institutional minutes identify the type-2 diabetes model as CCSS first health AI model, report its initial scope as complete and record a scaling proposal.
Teletica reported more than one million records analyzed, 2,500 people identified as at risk, 1,800 contacted and 130 already diagnosed.
Evidence matrix
Sources and traceability
Sources state which claims they support, their origin and the date on which they were consulted.
- [1]CCSS Board minutes, session 9504
- Publisher
- Caja Costarricense de Seguro Social
- Source type
- Official primary source
- Published
- Consulted
Supports: existence · execution · AI technique · operational use · reported result
- [2]LIDIA, the artificial-intelligence program growing within CCSS
- Publisher
- Teletica
- Source type
- News media
- Published
- Consulted
Supports: existence · execution · AI technique · operational use · reported result
Open questions
- What institutional protocol documents LIDIA's clinical validation, human review and handling of false positives?
Documented relationships
Depends on
EDUS: Unified Digital Health RecordThe models are built using information held in EDUS.
Context
It counts as a verified pilot because institutional minutes confirm execution, technique and use of the model. Detailed figures come from the press and are presented as reported results, not an independent evaluation.
Related projects
Ecosystem and capabilitiesOperational
Verified:
EDUS: Unified Digital Health Record
CCSS digital infrastructure that centralizes clinical records and enables information exchange across facilities. EDUS is the data foundation used by separate initiatives such as LIDIA and the announced AIDA integration.
- Initiative type
- Digital infrastructure
- Operational start
Ecosystem and capabilitiesCompleted
Verified:
TEC-CCSS training program in medical AI
8-week course for medical, IT and administrative staff. Projects: anomaly detection in mammograms, analysis of neonatal eye imaging.
- Initiative type
- Capacity-building program
- First evidence
Initiatives under reviewPlanned
Verified:
AIDA: Intelligent Digital Assistant for Care
Intelligent Digital Assistant for Care announced by CCSS to support real-time clinical decisions and integrate with EDUS as part of its primary-care strengthening strategy.
- Initiative type
- AI component
- Announcement