# mwasifanwar/medvision-ai architecture

> Architecture diagram of the GitHub repository mwasifanwar/medvision-ai, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2025-11-01). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/mwasifanwar/medvision-ai
- Repository: https://github.com/mwasifanwar/medvision-ai

## Overview

##### Project Type and Purpose
- This is a modular, pipeline-oriented medical imaging diagnostic assistant written in Python.
- Purpose: Ingest DICOM images, preprocess and normalize them, run modality-specific AI analyzers, fuse multi-modal evidence, correlate findings with clinical context, and generate structured reports.

##### File Structure Analysis
- main.py
– Application entry point (likely a Streamlit app or CLI orchestrator).
- core/
– dicom_processor.py: DICOM loading, decoding, metadata extraction, normalization.
– model_manager.py: Downloading, versioning, and loading pre-trained models.
– predictor.py: Dispatches to modality-specific AI analyzers and fusion engine.
– clinical_correlator.py: Combines image findings with patient context for risk stratification.
– report_generator.py: Summarizes findings, grades severity, produces recommendations.
- utils/
– config.py: Parameter definitions (thresholds, file paths).
– medical_metrics.py: Dice loss, uncertainty quantification, other domain metrics.
– visualization.py: Plotting, overlay of masks, confidence heatmaps.
- requirements.txt: Core dependencies (PyTorch, MONAI, SimpleITK, Streamlit).

##### Insights from README
- Multi-stage diagnostic pipeline with stages: DICOM Processor → Preprocessor → X-Ray/MRI/CT Analyzers → Fusion Engine → Clinical Correlator → Report Generator → Quality Assurance.
- Key technologies: PyTorch + MONAI for models, SimpleITK/OpenCV for image ops, Streamlit for UI, Docker for deployment, PACS integration for DICOM storage.

## Diagram (Mermaid)

```mermaid
flowchart TD
    %% Pipeline stages
    subgraph "Core Pipeline"
        direction TB
        UI["User Interface\n(Streamlit/CLI)"]:::corePipeline
        DI["DICOM Ingestion Layer"]:::corePipeline
        PP["Preprocessing Layer"]:::corePipeline
        AA["AI Analysis Engines\n(X-Ray/MRI/CT)"]:::corePipeline
        FE["Fusion Engine"]:::corePipeline
        CC["Clinical Correlator"]:::corePipeline
        RG["Report Generator"]:::corePipeline
    end

    %% Quality Assurance
    QA["Quality Assurance Module\n(Uncertainty & Audit)"]:::utils

    %% Utility Services
    subgraph "Utility Services"
        direction TB
        CFG["Configuration"]:::utils
        MET["Medical Metrics"]:::utils
        VIS["Visualization"]:::utils
    end

    %% External Services
    subgraph "External Services/Data Stores"
        direction TB
        MS["Model Repository (S3/local)"]:::external
        PACS["DICOM PACS Server"]:::external
        CDB["Clinical Database"]:::external
        LOG["Logging/Monitoring"]:::external
    end

    %% Data Flow
    UI -->|"ingest request"| DI
    DI -->|"raw DICOM → tensor"| PP
    PP -->|"normalized data"| AA
    AA -->|"inference results"| FE
    FE -->|"fused findings"| CC
    CC -->|"contextualized findings"| RG
    RG -->|"report output"| UI

    %% Model Manager connection
    MM["Model Management"]:::corePipeline
    MM -->|"load models"| AA
    AA -->|"model versions"| MM

    %% QA taps
    AA -.->|"compute uncertainty"| QA
    CC -.->|"audit trace"| QA
    RG -.->|"confidence metrics"| QA

    %% Utils connections
    CFG -.-> DI
    CFG -.-> PP
    CFG -.-> AA
    CFG -.-> CC
    CFG -.-> RG

    MET -.-> QA
    MET -.-> AA
    MET -.-> CC

    VIS -.-> DI
    VIS -.-> PP
    VIS -.-> AA
    VIS -.-> RG

    %% External data sources
    PACS -->|"fetch DICOM"| DI
    MS -->|"model storage"| MM
    CDB -->|"patient metadata"| CC
    LOG -->|"logs & metrics"| QA

    %% Click Events
    click UI "https://github.com/mwasifanwar/medvision-ai/blob/main/main.py"
    click DI "https://github.com/mwasifanwar/medvision-ai/blob/main/core/dicom_processor.py"
    click PP "https://github.com/mwasifanwar/medvision-ai/blob/main/core/dicom_processor.py"
    click MM "https://github.com/mwasifanwar/medvision-ai/blob/main/core/model_manager.py"
    click AA "https://github.com/mwasifanwar/medvision-ai/blob/main/core/predictor.py"
    click FE "https://github.com/mwasifanwar/medvision-ai/blob/main/core/predictor.py"
    click CC "https://github.com/mwasifanwar/medvision-ai/blob/main/core/clinical_correlator.py"
    click RG "https://github.com/mwasifanwar/medvision-ai/blob/main/core/report_generator.py"
    click QA "https://github.com/mwasifanwar/medvision-ai/blob/main/utils/medical_metrics.py"
    click CFG "https://github.com/mwasifanwar/medvision-ai/blob/main/utils/config.py"
    click MET "https://github.com/mwasifanwar/medvision-ai/blob/main/utils/medical_metrics.py"
    click VIS "https://github.com/mwasifanwar/medvision-ai/blob/main/utils/visualization.py"

    %% Styles
    classDef corePipeline fill:#ADD8E6,stroke:#000,stroke-width:1px;
    classDef utils fill:#FFFACD,stroke:#000,stroke-width:1px,stroke-dasharray: 5 5;
    classDef external fill:#90EE90,stroke:#000,stroke-width:1px;
```
