Feature extraction failed

[helenabartels91]

I am trying to extract radiomic features in Jupyter notebook. This is my code for extraction. The images and masks are NRRD files extracted from 3D slicer.

path = (r'/Users/helen/Library/CloudStorage/OneDrive-Personal/Documents/Research/PAS Radiomics research/Radiomic feature extraction Pyradiomics') params = os.path.join(path, 'PAS.yaml')

inputCSV = os.path.join(path, 'PAS_superior_pathway_codes.csv') outputCSV = os.path.join(path, 'PAS_superior_Radiomic features.csv')

try: flists = pd.read_csv(inputCSV, sep=',').T except Exception: print('CSV READ FAILED') exit(-1)

extractor = featureextractor.RadiomicsFeatureExtractor(params)

results = pd.DataFrame()

imageFilepath = flists[entry]['Image'] maskFilepath = flists[entry]['Mask'] label = flists[entry].get('Label', None)

if str(label).isdigit(): label = int(label) else: label = None

featureVector = flists[entry]  # This is a pandas Series
featureVector['Image'] = os.path.basename(imageFilepath)
featureVector['Mask'] = os.path.basename(maskFilepath)

  result = pd.Series(extractor.execute(imageFilepath, maskFilepath, label))
  #featureVector = featureVector.append(result)
  featureVector = featureVector.concat((result), axis=0)
except Exception:
  print('FEATURE EXTRACTION FAILED:')

featureVector.name = entry
results = results.join(featureVector, how='outer')  # If feature extraction failed, results will be all NaN

results.T.to_csv(outputCSV,sep=';', index=False, na_rep='NaN')

I keep getting a message of "FEATURE EXTRACTION FAILED" Anyone see how to solve the problem?

Thank you

[JoostJM]

If you enable logging, PyRadiomics should output the reason why the feature extraction is failing. Try to use the commandline interface, this has the necessary code for enabling logging already there.