osirix.dcm_pix

Provides functionality for the images displayed in a 2D OsiriX viewer.

Example usage

import osirix
import matplotlib.pyplot as plt

viewer = osirix.frontmost_viewer()  # Raises GrpcException error if no viewer is available.
pix = viewer.cur_dcm()  # Get the currently displayed DCMPix object
plt.imshow(pix.image, cmap = "gray")  # Display the image data (pix.image is a 2D numpy array)
plt.show()

DCMPix

Bases: OsirixBase

image: NDArray property writable

The image data as a Numpy array.

If the image is RGB format, then the shape will be (rows, columns, 4), whereas if the image is greyscale format it will be shape (rows, columns) (see is_rgb property).

is_rgb: bool property

Is the image data red-green-blue? If False must be greyscale.

orientation: NDArray property

The orientation of the image.

NOTE: It can be more accurate to get the slice location by loading the dicom file via the source_file property (using pydicom for example), and using the ImageOrientationPatient tag.

origin: Tuple[float, float, float] property

The origin of the image (x, y, z).

NOTE: It can be more accurate to get the slice location by loading the dicom file via the source_file property (using pydicom for example), and using the ImagePositionPatient tag.

pixel_spacing: Tuple[float, float] property

The pixel spacing of the image (order: rows, columns)

shape: Tuple[int, int] property

The pixel shape of the image (order: rows, columns)

slice_location: float property

The slice location of the image.

NOTE: It can be more accurate to get the slice location by loading the dicom file via the source_file property (using pydicom for example), and using the ImagePositionPatient tag.

source_file: str property

The source file of the image on the host machine.

compute_roi(roi)

Compute some statistics from an ROI contained within the image.

Note that these are calculated internally by OsiriX. For more refinement on how statistics are calculated we suggest that it is better to create your own tools (based on the scipy.stats library for example).

Parameters:

Name	Type	Description	Default
roi	ROI	The region of interest from which to compute the statistics.	required

Returns:

Type	Description
Dict	A dictionary containing the following key-value pairs: "mean", "std", "min", "max", "skewness", "kurtosis"

convert_to_bw(mode=3)

Convert the image to greyscale.

Parameters:

Name	Type	Description	Default
mode	int	0 = use red channel, 1 = use green channel, 2 = use blue channel, and 3 = merge.	3

convert_to_rgb(mode=3)

Convert the image to RGB.

Parameters:

Name	Type	Description	Default
mode	int	0 = create red channel, 1 = create green channel, 2 = create blue channel, and 3 = create all channels.	3

get_map_from_roi(roi)

Create a mask from an input ROI based on the image.

Parameters:

Name	Type	Description	Default
roi	ROI	The ROI from which to extract the mask.	required

Returns:

Type	Description
NDArray	The mask as a 2-dimensional binary array with shape (rows, columns).

get_roi_values(roi)

Extract the pixel values within a region of interest.

Parameters:

Name	Type	Description	Default
roi	ROI	The ROI from which to extract values.	required

Returns:

Type	Description
NDArray	The row indices of the extracted values.
NDArray	The columns indices of the extracted values.
NDArray	The extracted pixel values.

image_obj()

The DicomImage instance from which the image was derived.

Returns:

Type	Description
DicomImage	The image instance.

series_obj()

The DicomSeries instance from which the image was derived.

Returns:

Type	Description
DicomSeries	The series instance.

study_obj()

The DicomStudy instance from which the image was derived.

Returns:

Type	Description
DicomStudy	The study instance.