EpiLyra
A cholera epidemic does not wait for paper forms to be digitized. EpiLyra is a frictionless, multimodal mobile app for Community Health Workers paired with a powerful AI “Clarity Engine.” It instantly standardizes messy field data (voice, text, or image) into WHO ICD-11 codes and uses geospatial algorithms to automatically trigger alerts for emerging outbreaks before they spread.
From noticing a symptom to launching an intervention.
Traditional surveillance depends on static digital forms — messy, slow, and often digitized long after an outbreak has taken hold. EpiLyra replaces them with AI-driven data standardization and spatial anomaly detection, drastically accelerating the public-health response.
The result is an antifragile approach that scales data processing without adding human personnel — and catches outbreaks before they spread.
Semantic Parser. An LLM-driven engine turns voice, text, or image — captured without typing — into standardized WHO ICD-11 codes.
Spatial Anomaly Trigger. Geospatial algorithms watch for emerging clusters and automatically raise alerts the moment a pattern breaks.
Built for the field and for headquarters.
Multimodal, No Typing
Community Health Workers log symptoms by voice, text, or image through a frictionless interface designed for the field — even where literacy or connectivity is low.
Instant ICD-11 Standardization
Messy field observations become clean, standardized WHO ICD-11 data automatically — no manual coding, no digitization backlog.
Automated Outbreak Alerts
Geospatial anomaly detection surfaces emerging outbreaks and pushes alerts to epidemiologists the moment a spatial pattern appears.
Offline-First & Antifragile
Speech-to-text and geospatial monitoring run offline-first in the most resource-deprived zones, syncing asynchronously when a link returns.
The research behind EpiLyra.
Four documents by Dr. Sohaib Khan — from foundational theory to the developer blueprint and ethical framework.
A Simplified AI Framework for Zero-Latency Surveillance
The foundational research: how replacing static digital forms with AI-driven standardization and spatial anomaly detection accelerates public-health response — an antifragile approach that catches outbreaks before they spread.
Field & Headquarters: A Dual-Perspective Guide
For Community Health Workers and headquarters epidemiologists alike — logging symptoms via voice or image in the field, and interpreting automated alerts and standardized ICD-11 data from the dashboard.
Building the Two-Tiered Clarity Engine
Exact specifications for the LLM-driven Semantic Parser and the Spatial Anomaly Trigger — integrating offline-first mobile capabilities with advanced speech-to-text and geospatial monitoring in low-latency, resource-deprived settings.
Antifragile Surveillance & Data Sovereignty
How EpiLyra protects vulnerable populations while expanding public-health capacity — locking down personally identifiable information locally, and aligning with IASC guidelines to remain ethically sound and operationally limitless.
“Even when the grid fails, our humanity — and our ability to help one another — should not.”
Most tech founders come from Silicon Valley. My perspective was forged in clinics, refugee contexts, and two decades of academic research into global health systems. As a Medical Doctor with a PhD in Public Health, I have spent my career studying how fragile systems break — and how to make them resilient.
