LIDIA — Predictive models in EDUS
Institution: Costa Rican Social Security Fund
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.
Verified results
Diabetes pilot at Clorito Picado Clinic: 1M+ patients analyzed, 2,500 identified as at-risk, 1,800 reached, 130 already diagnosed (model validation, 95% accuracy). Cost per model: ¢130M (~USD 250,000).
Context
Costa Rica’s only public preventive-medicine AI program. Built in-house by CCSS teams (physicians, engineers, statisticians) under EDUS director Manuel Rodríguez, with no external vendor. Three of the four models remain in phase zero due to budget constraints.
Official source
↗ https://www.teletica.com/salud/lidia-el-programa-de-inteligencia-artificial-que-crece-en-la-ccss-en-medio-de-un-dilema-etico_376322Related projects
TEC-CCSS training program in medical AI
Live8-week course for medical, IT and administrative staff. Projects: anomaly detection in mammograms, analysis of neonatal eye imaging.
Live since: 2025
EDUS with predictive AI
PlannedSingle Digital Health Record (live since 2010, Law 9162) with planned AI incorporation for predictive medicine.
AIDA — Intelligent Digital Assistant for Care
PilotAI assistant embedded in EDUS that supports health professionals in real time during consultation: suggests differential diagnoses, helps with chronic-disease follow-up, and enables asynchronous tele-consultation with specialists.
Live since: 2026