Skip to main content

Quantiphyse is a data viewer and analysis platform for volumetric medical imaging data

Project description

Quantiphyse

Viewer for 3D/4D data and Pk modelling

Overview

This viewer provides tools for modelling and analysis of 3D/4D volumetric data, principally MRI data.

Core features:

  • Loading/Saving 3D/4D NIFTI files
  • Analysis tools including single/multiple voxel analysis and data comparison
  • Generic processing including smoothing, resampling, clustering

Features available via plugins

  • Registration, motion correction
  • Modelling tools for DCE, ASL, DSC and CEST MRI
  • Integration of selected FSL tools

See: http://quantiphyse.readthedocs.org/en/latest/ for full documentation.

Installation

If you already have a Python installation, Quantiphyse is available on PyPi::

pip install quantiphyse

Note that Cython and Numpy must already be installed in order to build the core extensions. We hope to be able to provide wheels for Windows and Mac soon to avoid this.

Major releases of Quantiphyse are available via the Oxford University Innovation Software Store (https://process.innovation.ox.ac.uk/software). If you are interested in commercial licensing you shold contact OUI in the first instance. The packages held by OUI have no external dependencies and can be installed on Windows, Mac and Linux.

License

Quantiphyse is available free under an academic (non-commercial) license. See the LICENSE file for full details, and contact OUI if interested in commercial licensing.

Running from source code (for developers)

Running from source is recommended only if your are interested in developing the software further.

  1. Install the dependencies:

The list of Python dependencies is in requirements.txt

For example:

pip install -r requirements.txt
  1. Build extensions

python setup.py build_ext --inplace

  1. Run from source directory

python qp.py

Packaging

The scripts packaging/build.py is used to build a frozen distribution package in the form of a compressed archive (tar.gz or .zip) and a platform-dependent package (deb, msi or dpg). It should run autonomously, however you may need to input the sudo password on Linux in order to build a deb package.

The --snapshot option removes the version number from package filenames so you can provided them for download without having to change the link URLs.

The --maxi option builds a package which includes selected plugins, assuming these are downloaded

OSX

Installing from source on OSX is not fun. The major issue is QT since the required version (4.8) is deprecated and hard to install properly.

I have had most success using Anaconda and installing PySide using the conda tool. This should bring in the appropriate version of QT.

To Do list

Issue tracker

Current issues can be viewed on the GitHub issue tracker (https://github.com/ibme-qubic/quantiphyse/issues)

Roadmap

v0.6 (Released June 2018)

  • ASL tools first version (preprocess, model fit, calibration, multiphase)
  • Improved viewer (full resolution, aligned)

v0.8 (Target Mar 2019)

  • Integration of selected FSL tools (FLIRT, FAST, BET, FSL_ANAT?) DONE
  • Improved registration support (apply transform) DONE NEEDS TESTING
  • Improved manual data alignment tools PART DONE
  • Improved ASL tools based on oxasl (inc. ENABLE, VEASL, DEBLUR) PART DONE
  • Fabber T1 (integrate with existing T1 widget?) PART DONE
  • Fabber DCE (integrate with existing DCE widget?) PART DONE
  • DSC widget PART DONE
  • Improvements to ROI builder - working 'paint' tool PART DONE

v1.0 (Target June 2019)

  • Stable interface for QpWidget, QpData, Process
  • Otherwise no firm plans yet - selection from 'Vague plans' below

Vague Plans for Future

  • Python 3 (not fully tested currently but should be mostly OK)

  • Refactoring of view classes

    • This is a mess at the moment. Need all view options to be stored as metadata and cleaner separation between the ImageView widget and the individual OrthoView widgets.
  • MoCo/Registration

    • Bartek's MC method
  • 3D view

    • Probably not that useful but fun and may be easy(?) with vispy. Reliant on good refactoring of ImageView
    • Application to surfaces (Tom K?)
  • Add Jola's texture analysis which sounds cool, whatever it is

  • PK modelling validation

    • QIBA in progress
    • QIN
  • Simplify/rewrite generic Fabber interface

  • Improve memory usage by swapping out data which are not being displayed?

  • All widgets which process within ROI should work with the subimage within the bounding box of the ROI, not the whole image.

    • Supervoxels does this already with great performance improvement.
  • Support other file formats using NIBABEL.

    • DICOM conversion included where DCMSTACK is available
  • Add semiquantitative measures

    • Area under the curve
    • Enhancing fraction
  • Simulation tools

    • Motion simulation DONE
    • Add noise DONE
    • Fabber test data
    • 'Simulated brain'

Migration to PySide2

  • The current implementation uses PySide which is based on Qt4
  • Update to PySide2 when released which uses Qt5
  • Will provide support for HiDPI screens and proper scaling in OSx
  • PyQtgraph is currently the stumbling block as release version does not support Pyside2
  • Current git version has PySide2 modifications but not yet tested
  • Consider move to VisPy if this does not come to fruition

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

quantiphyse-0.7.7.tar.gz (1.5 MB view hashes)

Uploaded Source

Built Distributions

quantiphyse-0.7.7-cp34-cp34m-macosx_10_6_x86_64.whl (2.0 MB view hashes)

Uploaded CPython 3.4m macOS 10.6+ x86-64

quantiphyse-0.7.7-cp27-cp27m-win_amd64.whl (2.4 MB view hashes)

Uploaded CPython 2.7m Windows x86-64

quantiphyse-0.7.7-cp27-cp27m-macosx_10_7_x86_64.whl (2.0 MB view hashes)

Uploaded CPython 2.7m macOS 10.7+ x86-64

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page