Nifti output of autopipeline has different dimensions that original DICOM

[strixy16]

I ran autopipeline on a CT image from the TCIA LIDC-IDRI dataset and it changed the image dimensions.

I tried this specifically patient LIDC-IDRI-0001 with the CT series ID 1.3.6.1.4.1.14519.5.2.1.6279.6001.179049373636438705059720603192

When I load the DICOM image with read_dicom_series from imgtools.io, I get this:

# DICOM CT Loading
from imgtools.io import read_dicom_series

# Path to CT image directory
path_to_dicom_directory = '../../rawdata/LIDC-IDRI/images/LIDC-IDRI-0001/01-01-2000-NA-NA-30178/3000566.000000-NA-03192'

# load the CT using med-imagetools function 
ct_dicom = read_dicom_series(path_to_dicom_directory)

# Should return a SimpleITK image object
print("DICOM CT loaded:")
print("Type: ", type(ct_dicom))
print("Shape: ", ct_dicom.GetSize())

DICOM CT loaded:
Type:  <class 'SimpleITK.SimpleITK.Image'>
Shape:  (512, 512, 133)

but if I load the converted NIfTI using Nibabel, I get the following:

import nibabel as nib

# Path to single CT nifti file
path_to_ct_nifti = "../../procdata/mit_test/0_LIDC-IDRI-0001/CT/CT.nii.gz"

# Load in the image as a nibabel object
ct_nifti = nib.load(path_to_ct_nifti)

print("NIfTI image loaded:")
print("Type: ", type(ct_nifti))
print("Shape: ", ct_nifti.shape)

NIfTI image loaded:
Type:  <class 'nibabel.nifti1.Nifti1Image'>
Shape:  (360, 360, 133)

If I convert the DICOM image by loading it in with SimpleITK as above and then saving it with sitk.WriteImage, the dimensions don't change.