Skip to content

NeuronSize app

NeuronSize_app

Description

Interactive graphical interface to estimate the optimal neuron size for inter-session alignment and initialization in CNMF-E processing within the CaliAli pipeline. This tool processes a single representative frame to generate neuron-enhanced images across a range of filter sizes, allowing manual selection of the most appropriate size.

Function Inputs:
Parameter Name Type Description
lim Vector (Optional) Range of neuron filter radii to evaluate, in pixels. Default: 1:0.5:5
Function Outputs:
Output Name Type Description
r Scalar Selected neuron size (Gaussian filter sigma) in pixels
Example usage:
r = NeuronSize_app();          % Use default neuron size range
r = NeuronSize_app(1:0.5:10);   % Custom size range

The selected frame is chosen to highlight neuron contrast.

  • If the file name contains _mc, the maximum projection across time is used.
  • Otherwise, the frame with the global maximum intensity is selected.
    The goal is to highlight neuron bodies clearly across size scales.

Neuron Size Selection Interface

The app opens a window showing neuron-enhanced images computed using various filter sizes (gSig).
In the interface: - Use the slider to select the size at which neurons appear best defined. - The images update interactively to reflect the enhancement level at each size.

Once satisfied, press Ok! to confirm your selection.
The chosen value will be printed in the MATLAB console:

The chosen Neuron size is [2.50]
Internally, the function performs the following:
  • Loads the video using CaliAli_load.
  • Chooses a representative frame based on motion correction status.
  • Applies border removal using remove_borders.
  • Enhances neuron visibility across sizes using neuron_stack.
  • Displays the results in a GUI using NeuronSize_app_in for manual selection.