src.algorithms.segment package

Submodules

src.algorithms.segment.trained_model module

algorithms.segment.trained_model

An API for a trained segmentation model to predict nodule boundaries and descriptive statistics.

src.algorithms.segment.trained_model.calculate_volume(segment_path, centroids)

Calculates tumor volume from pixel masks

Parameters:

• segment_path (str) – A path to the serialized binary mask for each centroid
• centroids (list[dict]) –

  A list of centroids of the form::

  {‘x’: int,

  ‘y’: int, ‘z’: int}

Returns:

List of volumes per centroid

Return type:

list[float]

src.algorithms.segment.trained_model.predict(dicom_path, centroids)

Predicts nodule boundaries.

Given a pth to a DICOM image and a list of centroids

1. load the segmentation model from its serialized state
2. pre-process the dicom data into whatever format the segmentation model expects
3. for each pixel create an indicator 0 or 1 of if the pixel is cancerous
4. write this binary mask to disk, and return the path to the mask

Parameters:

• dicom_path (str) – a path to a DICOM directory
• centroids (list[dict]) –

  A list of centroids of the form::

  {‘x’: int,

  ‘y’: int, ‘z’: int}

Returns:

Dictionary containing path to serialized binary masks and

volumes per centroid with form:: {‘binary_mask_path’: str,

‘volumes’: list[float]}

Return type:

dict

Module contents