ptufile 2024.4.24

Read PicoQuant PTU and related files

Ptufile is a Python library to read image and metadata from PicoQuant PTU and related files: PHU, PCK, PCO, PFS, PUS, and PQRES. PTU files contain time correlated single photon counting (TCSPC) measurement data and instrumentation parameters.

Author:

Christoph Gohlke

License:

BSD 3-Clause

Version:

2024.4.24

DOI:

10.5281/zenodo.10120021

Quickstart

Install the ptufile package and all dependencies from the Python Package Index:

python -m pip install -U ptufile[all]

See Examples for using the programming interface.

Source code and support are available on GitHub.

Requirements

This revision was tested with the following requirements and dependencies (other versions may work):

• CPython 3.9.13, 3.10.11, 3.11.9, 3.12.3 (64-bit)

• Numpy 1.26.4

• Xarray 2024.3.0 (recommended)

• Matplotlib 3.8.4 (optional)

• Tifffile 2024.4.24 (optional)

• Numcodecs 0.12.1 (optional)

• Cython 3.0.10 (build)

Revisions

2024.4.24

• Build wheels with numpy 2.

2024.2.20

• Change definition of PtuFile.frequency (breaking).

• Add option to specify number of bins returned by decode_histogram.

• Add option to return histograms of one period.

2024.2.15

• Add PtuFile.scanner property.

• Add numcodecs compatible PTU codec.

2024.2.8

• Support sinusoidal scanning correction.

2024.2.2

• Change positive dtime parameter from index to size (breaking).

• Fix segfault with ImgHdr_TimePerPixel = 0.

• Rename MultiHarp to Generic conforming with changes in PicoQuant reference.

2023.11.16

• Fix empty line when first record is start marker.

2023.11.13

• Change image histogram dimension order to TYXCH (breaking).

• Change frame start to start of first line in frame (breaking).

• Improve trimming of incomplete frames (breaking).

• Remove trim_dtime option (breaking).

• Fix selection handling in PtuFile.decode_image.

• Add option to trim T, C, and H axes of image histograms.

• Add option to decode histograms to memory-mapped or user-provided arrays.

• Add __getitem__ interface to image histogram.

2023.11.1

• Initial alpha release.

Notes

The Chan Zuckerberg Initiative financially supported the development of this library.

PicoQuant GmbH is a manufacturer of photonic components and instruments.

The PicoQuant unified file formats are documented at the PicoQuant-Time-Tagged-File-Format-Demos.

The following features are currently not implemented: PT2 and PT3 files, decoding images from T2 formats, bidirectional scanning, and deprecated image reconstruction.

Other Python modules for reading PicoQuant files are:

• Read_PTU.py

• readPTU_FLIM

• PyPTU

• tttrlib

• picoquantio

• ptuparser

• phconvert

• trattoria (wrapper of trattoria-core and tttr-toolbox)

• napari-flim-phasor-plotter

Examples

Read properties and tags from any type of PicoQuant unified tagged file:

>>> pq = PqFile('tests/Settings.pfs')
>>> pq.magic
<PqFileMagic.PFS: ...>
>>> pq.guid
UUID('86d428e2-cb0b-4964-996c-04456ba6be7b')
>>> pq.tags
{...'CreatorSW_Name': 'SymPhoTime 64', 'CreatorSW_Version': '2.1'...}
>>> pq.close()

Read metadata from a PicoQuant PTU FLIM file:

>>> ptu = PtuFile('tests/FLIM.ptu')
>>> ptu.magic
<PqFileMagic.PTU: ...>
>>> ptu.type
<PtuRecordType.PicoHarpT3: 66307>
>>> ptu.measurement_mode
<PtuMeasurementMode.T3: 3>
>>> ptu.measurement_submode
<PtuMeasurementSubMode.IMAGE: 3>

Decode TTTR records from the PTU file to numpy.recarray.

>>> decoded = ptu.decode_records()

Get global times of frame changes from markers:

>>> decoded['time'][(decoded['marker'] & ptu.frame_change_mask) > 0]
array([1571185680], dtype=uint64)

Decode TTTR records to overall delay-time histograms per channel:

>>> ptu.decode_histogram(dtype='uint8')
array([[ 5,  7,  7, ..., 10,  9,  2]], dtype=uint8)

Get information about the FLIM image histogram in the PTU file:

>>> ptu.shape
(1, 256, 256, 2, 3126)
>>> ptu.dims
('T', 'Y', 'X', 'C', 'H')
>>> ptu.coords
{'T': ..., 'Y': ..., 'X': ..., 'H': ...}
>>> ptu.dtype
dtype('uint16')

Decode parts of the image histogram to numpy.ndarray using slice notation. Slice step sizes define binning, -1 being used to integrate along axis:

>>> ptu[:, ..., 0, ::-1]
array([[[103, ..., 38],
        ...
        [ 47, ..., 30]]], dtype=uint16)

Alternatively, decode the first channel and integrate all histogram bins to a xarray.DataArray, keeping reduced axes:

>>> ptu.decode_image(channel=0, dtime=-1, asxarray=True)
<xarray.DataArray (T: 1, Y: 256, X: 256, C: 1, H: 1)> ...
array([[[[[103]],
           ...
         [[ 30]]]]], dtype=uint16)
Coordinates:
  * T        (T) float64... 0.05625
  * Y        (Y) float64... -0.0001304 ... 0.0001294
  * X        (X) float64... -0.0001304 ... 0.0001294
  * H        (H) float64... 0.0
Dimensions without coordinates: C
Attributes...
    frequency:      19999200.0
...
>>> ptu.close()

Preview the image and metadata in a PTU file from the console:

python -m ptufile tests/FLIM.ptu