Skip to content

A Python package for detecting somas in ExaSPIM image datasets. Includes tools for image preprocessing and deep learning-based classification and training.

License

Notifications You must be signed in to change notification settings

AllenNeuralDynamics/aind-exaspim-soma-detection

Repository files navigation

aind-exaspim-soma-detection

License Code Style semantic-release: angular Interrogate Coverage Python

Overview

To do...

pipeline
Figure: Visualization of soma detection pipeline. See "Inference" section for description of each step.

Add code block that shows how to run code...

Add image of what the code prints out...

Inference

Step 1: Proposal Generation

The objective of this step is to generate initial proposals for potential soma locations by detecting blob-like structures in the image. Our proposal generation algorithm consists the following steps:

a. Smooth image with Gaussian filter to reduce false positives.

b. Laplacian of Gaussian (LoG) to enhance regions where the gradient changes rapidly, then apply a non-linear maximum filter.

c. Generate initial set of proposals by detecting local maximas.

d. Shift each proposal to the brightest voxel in its neighborhood. If the brightness is below a threshold, reject the proposal.

proposals
Figure: Example of proposals generated across a large region.

Step 2: Filter Proposals with Heuristics

To do...

Step 3: Classify Proposals with Convolutional Neural Network (CNN)

To do...

Train Classification Model

To do...

Installation

To use the software, in the root directory, run

pip install -e .

About

A Python package for detecting somas in ExaSPIM image datasets. Includes tools for image preprocessing and deep learning-based classification and training.

Resources

License

Stars

Watchers

Forks

Packages

No packages published