Python porting of Ito et al., "Low computational-cost cell detection method for calcium imaging data", Neuroscience Research, 2022

Quickstart

1. clone this repository

git clone https://github.com/magnetizedCell/lccd-python.git

2. install required packeages.

pip3 install -r lccd-python/requirements.txt

3. • execute lccd in command line

     cd lccd-python
     python3 lccd_python/lccd.py --conf config/config.json --src profiles/src.npy --dst out.dump

     Here source file is a numpy array file that have dimesion (X, Y, T).
   • use lccd as module
     read profiles/test_lccd.ipynb

TODO

• [x]Compare result with official implementation
  Especially roi_integration -- seems ok
• []Make commnad line src file loading more flexible
• []Make variable and function names pythonic
• []Add GPU mode (see tests/profile_blob_detector.ipynb)
• []Add lazy frame loading for memory-efficient computation
• [x]Try sparse matrix in roi_integration: specify sparse:true in config.config.json