dicom2nifti Documentation

By default, dicom2nifti does not convert single slice images. You can explicitly enable this via the CLI or by modifying settings in Python.

# CLI approach
dicom2nifti -S input_directory output_directory

import dicom2nifti
import dicom2nifti.settings as settings

settings.disable_validate_slicecount()
dicom2nifti.convert_directory(dicom_directory, output_folder)

Support for inconsistent slice increments is disabled by default. To allow these images, you should enable resampling to avoid geometric distortions. You must configure the padding value and spline interpolation order.

# CLI approach
dicom2nifti -I -r -o 1 -p -1000 input_directory output_directory

import dicom2nifti
import dicom2nifti.settings as settings

settings.disable_validate_slice_increment()
settings.enable_resampling()
settings.set_resample_spline_interpolation_order(1)
settings.set_resample_padding(-1000)

dicom2nifti.convert_directory(dicom_directory, output_folder)

By default, dicom2nifti validates image orthogonality and disables support for gantry tilted CT. To process this data, you must disable validation and optionally enable resampling to create an orthogonal NIfTI.

Important: When using orthogonal resampling, the output NIfTI will always be reoriented. You should configure the padding value and spline interpolation order.

# CLI approach
dicom2nifti -G -r -o 1 -p -1000 input_directory output_directory

import dicom2nifti
import dicom2nifti.settings as settings

settings.disable_validate_orthogonal()
settings.enable_resampling()
settings.set_resample_spline_interpolation_order(1)
settings.set_resample_padding(-1000)

dicom2nifti.convert_directory(dicom_directory, output_folder)

To update the package to the latest version, use the following commands depending on your package manager.

# Using conda
conda update dicom2nifti

# Using pip
pip install dicom2nifti --upgrade

You can install dicom2nifti using either conda or pip.

Note on compressed formats: To support compressed DICOM formats like JPEG and JPEG2000, you must also install gdcm in combination with pydicom.

# Using conda
conda install -c conda-forge dicom2nifti

# Using pip
pip install dicom2nifti

You can integrate dicom2nifti into your Python workflows using two primary methods:

1. convert_directory: Converts an entire directory containing multiple DICOM series into NIfTI files.
2. dicom_series_to_nifti: Converts a directory containing exactly one DICOM series into a single NIfTI file.

import dicom2nifti

# Convert a directory with multiple series
dicom2nifti.convert_directory(dicom_directory, output_folder, compression=True, reorient=True)

# Convert a directory with only 1 series to 1 nifti file
dicom2nifti.dicom_series_to_nifti(original_dicom_directory, output_file, reorient_nifti=True)

The dicom2nifti CLI tool allows you to convert DICOM directories to NIfTI format.

Usage: dicom2nifti [options] input_directory output_directory

Available Flags:

• -h: Show help
• -G: Allow gantry tilted CT
• -r: Enable resampling
• -o RESAMPLE_ORDER: Set resampling spline interpolation order
• -p RESAMPLE_PADDING: Set resampling padding value
• -M: (Not explicitly detailed in help summary, but available)
• -C: (Not explicitly detailed in help summary, but available)
• -R: (Not explicitly detailed in help summary, but available)
• -I: Allow inconsistent slice increments
• -S: Allow single slice images

dicom2nifti [-h] [-G] [-r] [-o RESAMPLE_ORDER] [-p RESAMPLE_PADDING] [-M] [-C] [-R] input_directory output_directory

# Example: Allow gantry tilted CT with resampling
dicom2nifti -G -r -o 1 -p -1000 input_directory output_directory