PanImg Models

Data models for medical image processing and management.

Overview

This module provides Pydantic dataclasses and enums for handling medical imaging data, including support for various image formats, metadata, and processing results.

Core Components

Enums

ImageBuilderOptions: Supported image builder formats

• MHD, NIFTI, NRRD, DICOM, TIFF, OCT, FALLBACK

PostProcessorOptions: Post-processing formats

• DZI (Deep Zoom Image)

ColorSpace: Image color spaces

• GRAY, RGB, RGBA, YCBCR

ImageType: Output image types

• MHD, TIFF, DZI

EyeChoice: Eye selection for ophthalmology images

• OD (Oculus Dexter - right eye)
• OS (Oculus Sinister - left eye)
• U (Unknown)
• NA (Not Applicable)

Data Models

PanImg: Immutable image metadata

PanImg(
    pk=uuid4(),
    name="scan.mhd",
    width=512,
    height=512,
    depth=100,
    voxel_width_mm=0.5,
    voxel_height_mm=0.5,
    voxel_depth_mm=1.0,
    color_space=ColorSpace.GRAY,
    eye_choice=EyeChoice.OD,
    segments=frozenset({1, 2, 3})
)

PanImgFile: Image file reference

PanImgFile(
    image_id=uuid4(),
    image_type=ImageType.DZI,
    file=Path("/data/image.dzi"),
    directory=Path("/data/image_files")
)

PanImgResult: Image processing results

PanImgResult(
    new_images={img1, img2},
    new_image_files={file1, file2},
    consumed_files={Path("/input/scan.mhd")},
    file_errors={Path("/bad.mhd"): ["Invalid header"]}
)

PostProcessorResult: Post-processing output

PostProcessorResult(
    new_image_files={dzi_file}
)

Constants

• MAXIMUM_SEGMENTS_LENGTH = 64: Max segments for int8/uint8 data
• ITK_COLOR_SPACE_MAP: Maps ITK channel counts to ColorSpace

Testing

Run tests with pytest:

pytest test_module.py -v

Requirements

• Python 3.10+
• pydantic
• pytest (for testing)