Mitoclass
MitoClass is a napari plugin for classifying mitochondrial morphology from microscopy images: it allows preprocessing data, predicting classes (connected, fragmented, intermediate), visualizing overlays and 3D summaries, and managing a prediction history.
MitoClass
1 Overview
MitoClass is a napari plugin for the qualitative assessment of mitochondrial network morphology. Inference is patch‑wise: each 2‑D patch—obtained from a maximum‑intensity projection of 3‑D stacks—is classified as connected, fragmented, or intermediate.
2 Key features
| Module | Description |
|---|---|
| Patch‑based inference | Analyse an image folder or the active napari layer. |
| RGBA heatmaps | Overlay prediction maps as semi‑transparent layers in napari. |
| Global statistics | Compute the proportion of pixels assigned to each morphology and identify the dominant class. |
| 3‑D graph | Interactive Plotly scatter plot of connected / fragmented / intermediate proportions per image. |
3 Requirements
- Python ≥ 3.10
- OS : Windows, Linux or macOS
- Hardware : CPU is sufficient; GPU (CUDA 11+) is recommended for large datasets
4 Installation
4.1 PyPI
pip install mitoclass
4.2 Reproducible conda environment
conda create -n mitoclass python=3.10
conda activate mitoclass
# (Optional) GPU acceleration
conda install -c conda-forge cudnn=8.9 cuda11.8 tensorflow
pip install mitoclass
Apple Silicon: install tensorflow-macos.
4.3 Pre‑trained model
Download the model (.h5) from
<>
5 Usage
5.1 Graphical interface
napari
-
Open Plugins → MitoClass.
-
Select the four required paths:
Field Purpose Input dir Folder of images to analyse ( .tif,.tiff,.stk,.png).Output dir Destination folder for CSV and graph files. Heatmaps dir Folder where heatmaps ( *_map.tif) will be written.Model file Pre‑trained Keras model ( .h5). -
Click Run inference. A progress bar tracks the number of processed images.
-
After completion:
- Show heatmaps adds the newly generated
*_map.tiflayers to napari. - Show 3D graph opens
graph3d.html, displaying the connected/fragmented/intermediate proportions.
- Show heatmaps adds the newly generated
Tip: Without an Input dir you may run Infer active layer; results are still saved to Output dir and Heatmaps dir.
5.2 Output structure
| Folder | File(s) | Content |
|---|---|---|
| Output dir | predictions.csv |
Pixel proportion for each class (connected, fragmented, intermediate) and the dominant morphology, one line per image. |
graph3d.html |
Interactive 3‑D Plotly graph of class proportions. | |
| Heatmaps dir | *_map.tif |
One RGBA heatmap per image, ready to overlay in napari. |
6 Licence
This project is released under the GNU GPL v3 licence. See the LICENSE file for details.
Version:
- 1.0.3
Last updated:
- 2025-09-17
First released:
- 2025-07-30
License:
- GNU GENERAL PUBLIC LICENSE ...
Operating system:
- Information not submitted
Requirements:
- numpy
- magicgui
- qtpy
- scikit-image
- tifffile
- tensorflow
- pandas
- scikit-learn
- plotly
- napari[all]
- napari[all]; extra == "all"
- tox; extra == "testing"
- pytest; extra == "testing"
- pytest-cov; extra == "testing"
- pytest-qt; extra == "testing"
- napari[qt]; extra == "testing"