search

FNNDSC / pl-fetal-brain-assessment

Quality Assessment tool for fetal brain MRIs, able to score each volume through a deep learning regression model. Developed using Python3 and Keras/Tensorflow framework.
MIT License
3 stars 1 forks source link

Hard-coded image dimensions are not generalized #4

Open jennydaman opened 3 years ago

jennydaman commented 3 years ago

https://github.com/FNNDSC/pl-fetal-brain-assessment/blob/04adb7d08ced030bcd670e05ddb7ca4a60dc69ff/fetal_brain_assessment/volume.py#L48-L49

Input is required to have very specific dimensions. Not sure about how dimensions of the image affect model evaluation.

jennydaman commented 1 week ago 
Traceback (most recent call last):
  File "/opt/conda/bin/fetal_brain_assessment", line 8, in <module>
    sys.exit(main())
  File "/opt/conda/lib/python3.8/site-packages/fetal_brain_assessment/__main__.py", line 6, in main
    chris_app.launch()
  File "/opt/conda/lib/python3.8/site-packages/chrisapp/base.py", line 462, in launch
    self.run(self.options)
  File "/opt/conda/lib/python3.8/site-packages/fetal_brain_assessment/fetal_brain_assessment.py", line 145, in run
    volumes = [Volume(f) for f in input_files]
  File "/opt/conda/lib/python3.8/site-packages/fetal_brain_assessment/fetal_brain_assessment.py", line 145, in <listcomp>
    volumes = [Volume(f) for f in input_files]
  File "/opt/conda/lib/python3.8/site-packages/fetal_brain_assessment/volume.py", line 51, in __init__
    pad[:data.shape[0], :data.shape[1], :data.shape[2]] = data
ValueError: could not broadcast input array from shape (77,70,67,1) into shape (77,70,60,1)