IAPEX (Encuéntrame)
Hybrid AI Ecosystem for Patient Identification
A centralized platform bridging the gap between unidentified hospital patients and searching families using a Hybrid Information Fusion engine (Facial + Textual) to maximize identification accuracy under strict privacy protocols.
The challenge
Challenge
Hospitals rely on isolated, manual protocols to register 'John Does,' creating a critical data disconnect with families who search blindly in morgues and ERs.
Solution
A privacy-focused ecosystem connecting institutional data with public queries through a secure biometric matching core, reducing identification time from days to seconds.
Ecosystem
Neural Core
The heart of IAPEX. Fuses FaceNet embeddings (Euclidean distance) with text filters to rank candidates, significantly reducing false positives compared to standard recognition.
The Mobile Client (Family Side)
Secure interface for families to input descriptions. It displays potential matches based on similarity scores, protecting patient privacy until verification.
