# arturia-pendragon-iris/hipas_av_segmentation architecture

> Architecture diagram of the GitHub repository arturia-pendragon-iris/hipas_av_segmentation, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2025-01-05). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/arturia-pendragon-iris/hipas_av_segmentation
- Repository: https://github.com/arturia-pendragon-iris/hipas_av_segmentation

## Overview

##### Main Components Identification

a) HiPaS (Core Segmentation Module)
b) Reconstruction Module
c) Analysis Module

##### Detailed Architecture Breakdown

A. Data Flow Layer
- Input: CT Scans (HDF5 format)
- Data Preprocessing: Normalization (-1000, 600) to (0, 1)
- Output: Segmented Arteries and Veins

B. Processing Components
##### HiPaS Core (Main Module)
- Training Pipeline
- Dataset Management (datasets/)
- Data Augmentation (augment/)
- 3D UNet Architecture (unet3d/)
- Model Configuration (train_config_segmentation.yaml)
- Prediction Pipeline (predict_av.py)

##### Reconstruction Module
- Custom Loss Functions (losses.py)
- Reconstruction Model (model.py)
- Reconstruction Process (reconstruct.py)

##### Analysis Module
- Connectivity Analysis (connectivity.py)
- Distance Transform (distance_transform.py)
- Skeleton Analysis (get_skeleton.py)
- Surface Analysis (get_surface.py)

Use the following visual elements:
- Blue rectangles: Main modules
- Green rectangles: Sub-modules
- Yellow diamonds: Data processing steps
- Red hexagons: Model components

##### Key Relationships to Highlight

a) Training Flow:
**Raw CT Data → Data Preprocessing → Training Pipeline → Model Training → Trained Model**

b) Inference Flow:
**Input CT → Preprocessing → HiPaS Model → Segmentation → Reconstruction → Analysis**

##### Technical Details

- Input specifications: 512x512x512 normalized CT scans
- Model architecture: Modified 3D UNet
- Output format: Two-channel segmentation (arteries and veins)
- Processing time: ~2 minutes per CT volume

##### Implementation Notes

- The modular nature of the system
- Clear separation between training and inference pipelines
- Data transformation steps
- Integration points between modules

## Diagram (Mermaid)

```mermaid
flowchart TB
    subgraph Data_Flow
        Input[("CT Scans (HDF5)")]
        Preprocess["Data Preprocessing"]
        Output[("Segmented A/V")]
    end

    subgraph HiPaS_Core["HiPaS Core Module"]
        direction TB
        subgraph Training_Pipeline["Training Pipeline"]
            direction TB
            DatasetMgmt["Dataset Management"]:::component
            DataAug["Data Augmentation"]:::component
            UNet["3D UNet Architecture"]:::component
            Config["Model Configuration"]:::component
            Train["Training Process"]:::component
        end
        
        PredictPipe["Prediction Pipeline"]:::component
        CoreModel["Core Model"]:::component
    end

    subgraph Reconstruction["Reconstruction Module"]
        direction TB
        LossFn["Custom Loss Functions"]:::component
        ReconModel["Reconstruction Model"]:::component
        ReconProcess["Reconstruction Process"]:::component
        ReconTrain["Training Process"]:::component
    end

    subgraph Analysis["Analysis Module"]
        direction TB
        Connect["Connectivity Analysis"]:::component
        Distance["Distance Transform"]:::component
        Skeleton["Skeleton Analysis"]:::component
        Surface["Surface Analysis"]:::component
    end

    %% Data Flow Connections
    Input --> Preprocess
    Preprocess --> Training_Pipeline
    Preprocess --> PredictPipe
    PredictPipe --> Output
    Output --> Reconstruction
    Reconstruction --> Analysis

    %% Training Pipeline Connections
    DatasetMgmt --> DataAug
    DataAug --> UNet
    UNet --> Train
    Config --> Train
    Train --> CoreModel

    %% Reconstruction Connections
    LossFn --> ReconModel
    ReconModel --> ReconProcess
    ReconTrain --> ReconModel

    %% Analysis Module Connections
    Connect --> Distance
    Distance --> Skeleton
    Skeleton --> Surface

    %% Click Events
    click HiPaS_Core "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/tree/main/HiPaS"
    click PredictPipe "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/blob/main/HiPaS/predict_av.py"
    click CoreModel "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/blob/main/HiPaS/model.py"
    click Config "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/blob/main/HiPaS/train_config_segmentation.yaml"
    click DatasetMgmt "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/tree/main/HiPaS/train/datasets"
    click DataAug "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/tree/main/HiPaS/train/augment"
    click UNet "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/tree/main/HiPaS/train/unet3d"
    click Train "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/blob/main/HiPaS/train/train.py"
    click LossFn "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/blob/main/Reconstruction/losses.py"
    click ReconModel "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/blob/main/Reconstruction/model.py"
    click ReconProcess "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/blob/main/Reconstruction/reconstruct.py"
    click ReconTrain "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/blob/main/Reconstruction/train.py"
    click Connect "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/blob/main/analysis/connectivity.py"
    click Distance "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/blob/main/analysis/distance_transform.py"
    click Skeleton "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/blob/main/analysis/get_skeleton.py"
    click Surface "https://github.com/Arturia-Pendragon-Iris/HiPaS_AV_Segmentation/blob/main/analysis/get_surface.py"

    %% Styling
    classDef component fill:#a8d5ff,stroke:#333,stroke-width:2px
    classDef module fill:#ffcca8,stroke:#333,stroke-width:2px
    class HiPaS_Core,Reconstruction,Analysis module

    %% Legend
    subgraph Legend
        direction TB
        L1["Main Module"]:::module
        L2["Component"]:::component
        L3[("Data Storage")]
    end
```
