Python tools for handling Interfile and DICOM variants.

Usage

Converting DICOM to interfile

From command line:

pettools dcm2ifl INFILE.ptd OUTFILE

or in Python:

from petlink import InterfileCSA
iflcsa = InterfileCSA.from_file(in_file)
iflcsa.to_interfile(out_file)

Exploring Interfile data

from petlink import Interfile
ifl = Interfile('listmode.hl')

# access keys
ifl['patient orientation']  # e.g., HFS or head_in

# print all keys
print(ifl)

# access data as numpy array
ifl.get_data()

# parse date/time key pairs as Python datetime objects
ifl.get_datetime('study') - ifl.get_datetime('tracer injection')                                             
# -> datetime.timedelta(seconds=548)

Exploring ListMode data

from petlink import ListMode
lm = ListMode('listmode.ptd')  # or .IMA or .dcm+.bf

# get the duration
lm.duration   # time in ms

# raw data
lm.data  # 1D numpy array

# DICOM metadata
lm.dcm  # pydicom object

# Interfile header
lm.ifl  # Interfile object, see above

Indexing ListMode data by time

start_time_ms, end_time_ms = 1000, 2000

between_one_and_two_seconds = lm.tloc[start_time_ms:end_time_ms]

Installation

The package is configured with setuptools, and should integrate well with pip.
The simplest option is:

pip install --user git+https://github.com/ashgillman/petlink.git

Or to pull the repo locally an use:

pip install --user .

Nix users can run nix-shell or nix-env -if ..

Building

To use without installation, cython modules will need to be build in-place.

python3 setup.py build_ext --inplace

ashgillman/petlink