Add a command line interface for collageradiomics. #86

cli/collageradiomicscli.py

@@ -0,0 +1,93 @@
+#!usr/bin/env python
+# -*- coding: utf-8 -*-
+
+import click
+import os
+import sys
+import SimpleITK as sitk
+import csv
+from scipy import stats
+import numpy as np
+
+import collageradiomics
+
+@click.command()
+@click.option('-i', '--input', required=True, help='Path to an input image from which features will be extracted.')
+@click.option('-m', '--mask', required=True, help='Path to a mask that will be considered as binary. The highest pixel value will be considered as information and all other values will be considered outside the mask')
+@click.option('-o', '--outputfile', required=True, help='Path to the output CSV file.')
+@click.option('-v', '--verbose', default=True, help='Provides additional debug output.')
+@click.option('-d', '--dimensions', help='Optional number of dimensions upon which to run collage. Supported values are 2 and 3. If left out, we will default to the dimensionality of the image itself, which may not reflect expected behavior if the image has an alpha channel.', type=click.IntRange(2, 3, clamp=True))
+@click.option('-s', '--svdradius', default=5, help='SVD radius is used for the dominant angle calculation pixel radius. DEFAULTS to 5 and is suggested to remain at the default.')
+@click.option('-h', '--haralickwindow', default=-1, help='Number of pixels around each pixel used to calculate the haralick texture. DEFAULTS to svdradius * 2 - 1.')
+@click.option('-b', '--binsize', default=64, help='Number of bins to use while calculating the grey level cooccurence matrix. DEFAULTS to 64.')
+def run(input, mask, outputfile, verbose, dimensions, svdradius, haralickwindow, binsize):
+  """CoLlAGe captures subtle anisotropic differences in disease pathologies by measuring entropy of co-occurrences of voxel-level gradient orientations on imaging computed within a local neighborhood."""
+
+  image = sitk.ReadImage(input)
+  mask = sitk.ReadImage(mask)
+
+  image_array = sitk.GetArrayFromImage(image)
+  mask_array = sitk.GetArrayFromImage(mask)
+
+  # Remove any extra array dimensions if the user explicitly asks for 2D.
+  if dimensions == 2:
+    image_array = image_array[:,:,0]
+    mask_array  = mask_array [:,:,0]
+
+  collage = collageradiomics.Collage(
+    image_array, 
+    mask_array, 
+    svd_radius=svdradius, 
+    verbose_logging=verbose,
+    num_unique_angles=binsize)
+
+  collage.execute()
+
+  # Create a csv file at the passed in output file location.
+  with open(outputfile, 'w', newline='') as csv_output_file:
+    writer = csv.writer(csv_output_file)
+
+    # Write the columns.
+    writer.writerow(['FeatureName', 'Value'])
+    for feature in collageradiomics.HaralickFeature:
+      feature_output = collage.get_single_feature_output(feature)
+      if image_array.ndim == 2:
+        feature_output = feature_output[~np.isnan(feature_output)]
+
+        # NumPy supports median natively, we'll use that.
+        median = np.nanmedian(feature_output, axis=None)
+
+        # Use SciPy for kurtosis, variance, and skewness.
+        feature_stats = stats.describe(feature_output, axis=None)
+
+        # Write CSV row for current feature.
+        _write_csv_stats_row(writer, feature, median, feature_stats.skewness, feature_stats.kurtosis, feature_stats.variance)
+      else:
+        # Extract phi and theta angles.
+        feature_output_theta = feature_output[:,:,:,0]
+        feature_output_phi = feature_output[:,:,:,1]
+
+        # Remove NaN for stat calculations.
+        feature_output_theta = feature_output_theta[~np.isnan(feature_output_theta)]
+        feature_output_phi = feature_output_phi[~np.isnan(feature_output_phi)]
+
+        # NumPy supports median natively, we'll use that.
+        median_theta = np.nanmedian(feature_output_theta, axis=None)
+        median_phi = np.nanmedian(feature_output_phi, axis=None)
+
+        # Use SciPy for kurtosis, variance, and skewness.
+        feature_stats_theta = stats.describe(feature_output_theta.flatten(), axis=None)
+        feature_stats_phi = stats.describe(feature_output_phi.flatten(), axis=None)
+
+        # Write CSV rows for each angle.
+        _write_csv_stats_row(writer, feature, median_theta, feature_stats_theta.skewness, feature_stats_theta.kurtosis, feature_stats_theta.variance, 'Theta')
+        _write_csv_stats_row(writer, feature, median_phi, feature_stats_phi.skewness, feature_stats_phi.kurtosis, feature_stats_phi.variance, 'Phi')
+
+def _write_csv_stats_row(writer, feature, median, skewness, kurtosis, variance, suffix=''):
+  writer.writerow([f'Collage{feature.name}Median{suffix}', f'{median:.10f}'])
+  writer.writerow([f'Collage{feature.name}Skewness{suffix}', f'{skewness:.10f}'])
+  writer.writerow([f'Collage{feature.name}Kurtosis{suffix}', f'{kurtosis:.10f}'])
+  writer.writerow([f'Collage{feature.name}Variance{suffix}', f'{variance:.10f}'])
+
+if __name__ == '__main__':
+  run()

cli/setup.py

@@ -0,0 +1,33 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+from setuptools import setup
+
+setup(name='collageradiomicscli',
+      version='1.0.0',
+      description='Get Collage features from an image and a binary mask',
+      url='https://github.com/radxtools/collageradiomics',
+      python_requires='>=3.6',
+      author='Toth Technology',
+      author_email='[email protected]',
+      license='BSD-3-Clause',
+      zip_safe=False,
+      install_requires=[
+      'collageradiomics',
+      'setuptools>=47',
+      'SimpleITK==1.2.4',
+      'click'
+      ],
+      scripts=['collageradiomicscli.py'],
+      classifiers=[
+          'Intended Audience :: Science/Research',
+          'Intended Audience :: Developers',
+          'License :: OSI Approved :: BSD License',
+          'Operating System :: OS Independent',
+          'Operating System :: POSIX :: Linux',
+          'Operating System :: Microsoft :: Windows :: Windows 10',
+          'Operating System :: MacOS',
+          'Programming Language :: Python :: 3',
+          'Topic :: Scientific/Engineering :: Bio-Informatics',
+      ]
+      )