FLIMLib is a curve fitting library used for Fluorescent Lifetime Imaging or FLIM. It is developed by Paul Barber (UCL and KCL, London) and the Advanced Technology Group at the Oxford Institute for Radiation Oncology, University of Oxford, as well as the Laboratory for Optical and Computational Instrumentation at the University of Wisconsin-Madison. FLIMLib is used for FLIM functionality in the Time Resolved Imaging (TRI2) software, as well as in the FLIMJ plugin for ImageJ.

For exponential lifetime fitting there are three core algorithms within FLIMLib:

1. A triple integral method that does a very fast estimate of a single exponential lifetime component.
2. A Levenberg-Marquardt algorithm or LMA that uses an iterative, least-squares-minimization approach to generate a fit. This works with single, double and triple exponential models, as well as stretched exponential.
3. A Bayesian algorithm that combines evidence from each single photon to estimate lifetimes etc. It offers better performance with low photon counts.

There is also code to perform 'global' analysis over a number of signals simultaneously (e.g. over an image), where the lifetimes can be considered constant across the data set, but the amplitudes are allowed to vary for each signal. There is also a completely generic global analysis function. A third algorithm is available to perform phasor analysis.

In addition there is a non-negative linear least squares algorithm that is useful for spectral unmixing in combined spectral-lifetime imaging (SLIM).

The FLIMLib library code is written in C89 compatible C and is thread-safe for fitting multiple pixels concurrently. A Java interface (generated by SWIG is privided to call the library from Java code: FLIMLib.java provide a subset of function calls used by the FLIMJ plugin for ImageJ.

Additionally, there is wrapper code in FLIMLib.i to wrap the external functions in flimlib.def. This code generates swig wrapper files which enable you to call these functions from Java.

• FLIMLib wiki
• FLIMLib web site
• FLIMLib Doxygen docs

You need JDK, Maven, CMake, SWIG, and C and C++ toolchains (GCC on Linux, Command Line Tools or Xcode on macOS, Visual Studio (with C++ Desktop Development) on Windows) to be installed.

To build the library and standalone program using maven:

mvn clean install

1. Copy the executable to the test_files folder for convenience

   cp target/build/bin/flimlib-cmd ./test_files
   

2. Run the program with the test files

   cd ./test_files
   ./flimlib-cmd test.ini transient.dat
   

To depend on FLIMLib from Maven, simply copy the following to appropriate places in your pom.xml:

<properties>
  <flimlib.version>2.1.0</flimlib.version>
</properties>

<!-- FLIMLib Java interface -->
<dependency>
  <groupId>flimlib</groupId>
  <artifactId>flimlib</artifactId>
  <version>${flimlib.version}</version>
</dependency>
<!-- FLIMLib native binary -->
<dependency>
  <groupId>flimlib</groupId>
  <artifactId>flimlib</artifactId>
  <version>${flimlib.version}</version>
  <classifier>${scijava.natives.classifier}</classifier>
  <!-- Or one of the following if you would like to manually specify the binary platform -->
  <!-- <classifier>native-linux_64</classifier> -->
  <!-- <classifier>native-windows_64</classifier> -->
  <!-- <classifier>native-osx_64</classifier> -->
</dependency>

Note that the native binary is platform-dependent. So you may want to make sure that the <classifier> attribute is either automatically detected by the parent scijava pom (${scijava.natives.classifier}) or manually filled in to match your platform.

The Python API is a ctypes-based wrapper around a few of the library functions.

pip install flimlib

import flimlib

To get started, see the help (docstrings) for these functions:

• flimlib.GCI_marquardt_fitting_engine() (Levenberg-Marquardt)
• flimlib.GCI_triple_integral_fitting_engine() (RLD: rapid lifetime determination)
• flimlib.GCI_Phasor() (phasor analysis)