Full-Stack Clinical System · Flask & React
Diabetes Expert
Clinical decision support platform powered by a forward-chaining expert inference engine with certainty factors, explainable AI diagnostics, role-based access control (RBAC), and longitudinal patient telemetry.
Feature
Forward Chaining & Explainable AI
- Deterministic rule-based inference evaluating patient symptoms, laboratory metrics, and risk factors
- Transparent reasoning trace detailing fact extraction, active rule firings, and certainty factors
- Stratified clinical diagnosis output with personalized lifestyle and dietary guidance
Multi-Role Clinical Workflow (RBAC)
- Granular access controls separating Admin governance, Doctor consultations, and Patient telemetry
- Secure JWT session handling with refresh tokens, permission checks, and active token revocation
- Clinical review annotation tools allowing physicians to flag acute cases and override automated findings
Dynamic Knowledge Base & Rules Engine
- Interactive web-based rule editor enabling doctors to update diagnostic criteria without engineering support
- Immutable audit log tracking every rule modification, timestamp, and author for medical compliance
- Version snapshot history allowing rollback of rule sets to previously validated medical baselines
Patient Telemetry & Health Monitoring
- Longitudinal tracking of fasting blood glucose, HbA1c levels, and vital biomarker trends
- Interactive patient symptom consultation questionnaire assessing clinical indicators and nocturia severity
- Dynamic risk stratification generating personalized lifestyle interventions and clinical follow-up alerts
Technical
Deterministic rule inference engine generating transparent Explainable AI (XAI) reasoning traces
Robust role-based access control (RBAC) backed by PostgreSQL, Supabase, and token revocation mechanisms
Real-time rule editor enabling live knowledge-base updates without code deployment or downtime
Technologies Utilized
Python
Flask
SQLAlchemy
PostgreSQL
Supabase
React
Vite
Tailwind CSS
JWT
Docker