microVis

Interactive desktop GUI for visualizing microProfiler microscopy datasets — plate grids, multi-channel images, object annotation, and profiling data.

Requirements

• Python >= 3.10
• OS: Only Windows 11 (64-bit) was tested

Installation

Quick Install

conda create -n micro
conda activate micro

git clone https://github.com/soulong/microVis.git
cd microVis
pip install .

Quick Start

# Launch with folder selector
microvis

# Launch directly with a dataset
microvis "/path/to/Measurement 1"

Windows Desktop Shortcut

After installing in the micro conda environment, create a desktop shortcut that launches microVis without opening a console window:

microvis install-shortcut

This places a shortcut on the Desktop and in the Start Menu. The shortcut uses pythonw.exe from the micro environment for a clean, console-free launch.

Features

Well Plate Grid

Interactive scatter-plot grid visualizing well plates (24-, 96-, or 384-well). Color wells by any profiling column or merged metadata, with natural sorting, colormaps, and click-to-select.

Multi-Channel Image Compositing

Composite multi-channel microscopy images with per-channel controls:

• Enable/disable individual channels
• Assign display colors (green, red, magenta, blue, cyan, yellow, white)
• Adjust vmin/vmax per channel or use auto-range (percentile-based, per-channel)
• Apply contrast transforms: gamma correction, histogram equalization, invert

Full-Resolution Zoom

Ctrl + scroll on any thumbnail to zoom past native resolution. The app loads and composites the full-resolution image in the background and cross-fades it in when ready. Double-click to reset zoom. Middle-click drag to pan.

Object Overlay

Overlay segmentation masks and per-object profiling data onto images:

• Boundary contour visualization with per-object metric-based color-coded fills
• Per-object tooltips showing label ID and profiling values on hover
• Configurable colormap and alpha
• Colorbar showing the actual data range for the selected metric

Object Annotation

Drag-and-drop class labeling of segmented objects:

1. Create class names in the sidebar (e.g. "cell", "debris")
2. Drag objects directly from image thumbnails into class boxes
3. Objects are cropped and displayed as thumbnails inside each class box
4. Drag objects between class boxes to reclassify
5. Click an object thumbnail to remove it
6. Write all annotations to results.db as a new table

Object Export

Export cropped/masked objects to TIFF files with fine-grained control:

• Object range: export from currently displayed images, all images, or only annotated objects
• Channel mode: single-channel (one YX file per channel) or multi-channel (one CYX file per object)
• Max objects per image: randomly sample up to N objects per image (0 = no limit). In single-channel mode the file count is N × number of channels per image; in multi-channel mode it is N per image
• Annotated objects are exported into per-class subfolders

Metadata Integration

Import plate-shaped Excel metadata files, merge with profiling data, and use metadata columns for grid coloring or object overlay.

Profiling Data Browser

Browse profiling tables from results.db with:

• Sortable columns (natural sort for numeric values)
• Radio-button table selector
• Preview mode (top 20 rows) with total row count hint
• PyGwalker integration for interactive data exploration

Pixel Inspector

Click any image thumbnail to see per-channel pixel intensities at that coordinate.

Acknowledgements

microVis is a sibling package to microProfiler — a microscopy image preprocessing, segmentation, and profiling pipeline.

Citation

If you use microVis in your research, please cite the repository:

@software{microVis,
  author = {Hao He},
  title = {microVis: Interactive Desktop GUI for Microscopy Dataset Visualization},
  year = {2026},
  url = {https://github.com/soulong/microVis}
}

License

MIT