Technology & Architecture
Engineered for low-connectivity environments, offline resilience, multi-source observation capture, and human-verified decision support.
Capture diagnostic data wherever it originates.
AbbeDx unifies diverse diagnostic formats into standardized, interoperable observations without disrupting frontline laboratory workflows.
Manual Structured Entry
Numeric, qualitative & titre-based bench inputs
OCR-Assisted Capture
Digitisation of legacy bench worksheets & documents
AI-Assisted RDT Interpretation
Computer-vision image analysis for rapid test strips
Future Device Integration
Direct interfaces for benchtop analysers & medical devices
Diagnostic Observation Normalisation
Data standardized into clinical terms & structured metadata schema
MANDATORY HUMAN VERIFICATION
Every AI suggestion, OCR reading, and manual observation must be explicitly reviewed, validated, or edited by a qualified healthcare professional before final sign-off.
Built for healthcare environments where connectivity cannot be assumed.
Workflows continue uninterrupted during network outages or remote field deployments, guaranteeing data integrity at all times.
Multi-Device Access
Runs on smartphones, tablets, and desktop workstations in clinics and field settings.
Local Secure Storage
Encrypted local database stores diagnostic observations and queued actions offline.
Sync When Connected
Background synchronization triggers automatically upon internet or network reconnection.
Organisation Cloud
Consolidates multi-site data into central organizational and public health dashboards.
Responsible diagnostic AI by design.
Every automated pipeline in AbbeDx is constructed around clinical governance and professional oversight.
AI supports the observation. Healthcare professionals remain responsible for verification.
AbbeDx never replaces clinical judgment. Computer vision and optical OCR outputs serve as structured suggestions. A certified laboratory professional reviews the raw image or input, makes edits where necessary, and explicitly verifies the final record.
Human Verification
AI-generated observations and OCR extractions function exclusively as decision-support suggestions. A qualified healthcare professional must explicitly verify and sign off before finalization.
Auditability & Traceability
Complete log history tracking every raw acquisition image, model suggestion, user modification, and timestamped professional sign-off.
Privacy & Data Boundary
Strict role-based access control (RBAC), end-to-end local encryption, and organisation-partitioned patient record storage.
Standards Roadmap
Developed to support HL7 v2/v3, FHIR R4, ASTM E1394, and DHIS2 data exchange to connect seamlessly with national health platforms.
Designed to connect, not create another silo.
AbbeDx's architecture is being developed to support standards-based interoperability with laboratory analysers, hospital systems, and national health information infrastructure.