ER-Voice2Text Documentation

ER-Voice2Text is an advanced system for managing medical workflows in Emergency Departments, integrating automatic voice transcription, clinical entity extraction, and medical report generation.

System Overview

The system consists of:

• Django Backend: REST API for medical data management

• AI Services: Integration with NVIDIA NIM and Whisper models for transcription and analysis

• Database: SQLite for relational data, MongoDB for transcriptions and analysis

• React Frontend: User interface for doctors and healthcare operators

Key Features

• Real-time Audio Transcription: Using Whisper for accurate transcription

• Clinical Entity Extraction: LLM and NER for automatic identification of clinical data

• Complete Medical Workflow: From audio recording to final PDF report

• Medical Authentication: Login system for healthcare operators

• Analytics Dashboard: Statistics and visualizations for data analysis

Backend Architecture

The backend is organized in the following modules:

Backend Modules:

• core package
  • Subpackages
  • Submodules
  • core.admin module
  • core.apps module
  • core.models module
  • core.mongodb_models module
  • core.tests module
  • core.views module
• api package
  • Submodules
  • api.admin module
  • api.apps module
  • api.medical_workflow_views module
  • api.models module
  • api.serializers module
  • api.tests module
  • api.urls module
  • api.views module
• services package
  • TranscriptionService
  • ClinicalExtractionService
  • Submodules
  • services.clinical_extraction module
  • services.extraction module
  • services.mongodb_service module
  • services.ner_service module
  • services.nvidia_nim module
  • services.pdf_report module
  • services.transcription module
  • services.whisper_realtime module
  • services.whisper_service module
• medical_system package
  • Submodules
  • medical_system.asgi module
  • medical_system.settings module
  • medical_system.urls module
  • medical_system.wsgi module
• auth_views module
  • login_view()
  • logout_view()
  • health_check()
• manage module
  • main()

Django Models

The system uses Django models for managing:

• Doctor: Management of doctors and specializations

• Patient: Patient registry and clinical data

• Encounter: Emergency Department care episodes

• AudioTranscript: Audio transcriptions with metadata

• ClinicalData: Clinical data extracted from transcriptions

• ClinicalReport: Finalized medical reports

AI Services and Integration

The system integrates several services for intelligent analysis:

• NVIDIA NIM: Large Language Model for clinical entity extraction

• Whisper: High-precision speech-to-text transcription

• Text2NER: Named Entity Recognition for textual analysis

• MongoDB: Storage for transcriptions and unstructured data

Configuration and Deployment

For information on installation, configuration and deployment, consult the project README and setup documentation.

Indices and Tables

• Index

• Module Index

• Search Page