FluoSA (Fluorescence Signal Analyzer)

 

FluoSA inputs *.LIF and *.tif files, detects user-defined neural structures, and quantifies the fluorescence signal changes (frame-wise fluorescence intensity and the dF/F0) in these structures.

 

The outputs of FluoSA include:

1. An annotated video showing the detected neural structures:

   

2. Spreadsheets storing frame-wise fluorescence intensity in each detected neural structure:

   

3. Spreadsheets storing summary of fluorescence signal changes (dF/F0) in each detected neural structure:

   

 

Installation

1 Install FluoSA

python3 -m pip install FluoSA

2 Install CUDA toolkit v11.8

https://developer.nvidia.com/cuda-11-8-0-download-archive?target_os=Windows&target_arch=x86_64

3 Install Detectron2

python3 -m pip install 'git+https://github.com/facebookresearch/detectron2.git'

4 Install PyTorch 2.0.1

4.1 For Windows and Linux

4.1.1 If using GPU

python3 -m pip install torch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 --index-url https://download.pytorch.org/whl/cu118

4.1.2 CPU only

python3 -m pip install torch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 --index-url https://download.pytorch.org/whl/cpu

4.2 For Mac

python3 -m pip install torch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2

 

Usage

In your terminal / command prompt, type:

FluoSA

Then the user interface will be initiated: