omero-browser-qt 0.1.1

Reusable PyQt6 dialog for browsing and retrieving images from OMERO via ICE

omero-browser-qt

Reusable PyQt6 dialog for browsing and retrieving images from OMERO servers. Pixel data is fetched exclusively through the ICE transport layer so you always get the real stored values (not rendered RGB).

Features

• Login dialog with server-name history (credentials are never stored)
• QuPath-style browser — resizable dialog with:
  • Group / Owner filter combos
  • Lazy-loading Project → Dataset → Image tree
  • Image thumbnail preview (256 × 256)
  • Attribute table (dimensions, pixel sizes, acquisition date, …)
  • Name filter
• ICE pixel loading via RawPixelsStore
  • Full 5-D array fetch (getPlane)
  • Tile-based dask lazy loading for large / pyramidal images (getTile)
• Designed to be embedded in any PyQt6 application

Prerequisites: ZeroC ICE + omero-py

omero-py depends on ZeroC ICE, which cannot be built from source on modern Python. Pre-built wheels are available for Python 3.10 – 3.12 from Glencoe Software:

https://www.glencoesoftware.com/blog/2023/12/08/ice-binaries-for-omero.html

Install them before installing this package:

# 1. Install the pre-built ZeroC ICE wheel (pick your platform)
pip install https://github.com/glencoesoftware/zeroc-ice-py-linux-x86_64/releases/download/20240202/zeroc_ice-3.6.5-cp310-cp310-manylinux_2_28_x86_64.whl   # Linux / Python 3.10
pip install https://github.com/glencoesoftware/zeroc-ice-py-win-x86_64/releases/download/20240202/zeroc_ice-3.6.5-cp310-cp310-win_amd64.whl               # Windows / Python 3.10
pip install https://github.com/glencoesoftware/zeroc-ice-py-macosx-x86_64/releases/download/20240202/zeroc_ice-3.6.5-cp310-cp310-macosx_10_16_x86_64.whl   # macOS Intel / Python 3.10
# See the blog post above for cp311 / cp312 URLs and ARM variants.

# 2. Install omero-py (now that ICE is available)
pip install omero-py

Alternatively, use conda:

conda install -c conda-forge zeroc-ice omero-py

Installation

pip install omero-browser-qt

With the demo viewer

pip install "omero-browser-qt[viewer]"

Quick start

from PyQt6.QtWidgets import QApplication
from omero_browser_qt import OmeroGateway, LoginDialog, OmeroBrowserDialog

app = QApplication([])
gw = OmeroGateway()

# 1. Login (only server names are remembered)
if LoginDialog(gateway=gw).exec():

    # 2. Browse & select images
    dlg = OmeroBrowserDialog(gateway=gw)
    if dlg.exec():
        for img in dlg.get_selected_images():
            print(img.getName(), img.getId())

    gw.disconnect()

Loading pixel data

from omero_browser_qt import load_image_data, load_image_lazy, is_large_image

# For regular images — returns numpy arrays
result = load_image_data(image_wrapper)
volumes = result["images"]   # list of np.ndarray, one per channel, shape (Z, Y, X)
meta    = result["metadata"]

# For large / pyramidal images — returns dask arrays
if is_large_image(image_wrapper):
    result = load_image_lazy(image_wrapper)
    # result["images"] is a list (per channel) of lists (per resolution level) of dask arrays

Demo viewer

A stripped-down multi-channel image viewer is included as an example. It supports opening images from local files (via bioio) and from OMERO:

python examples/viewer_demo.py

Controls:

• Open File… — load a local OME-TIFF, ND2, CZI, etc.
• Open from OMERO… — launches the login + browser dialogs
• Projection — Slice / MIP (max intensity) / SUM
• Z slider — navigate slices (enabled in Slice mode)
• Lo% / Hi% — percentile-based contrast adjustment
• Channel buttons — toggle individual channels on/off
• Mouse wheel — zoom; middle-drag — pan

Publishing to PyPI

# 1. Install build tools
pip install build twine

# 2. Build the package
python -m build          # creates dist/omero_browser_qt-0.1.0.tar.gz + .whl

# 3. Validate
twine check dist/*

# 4. Create a PyPI account and API token
#    → https://pypi.org/manage/account/token/

# 5. Upload
twine upload dist/*
#    Username: __token__
#    Password: <your-api-token>

For subsequent releases: bump the version in pyproject.toml, rebuild, and re-upload.

License

MIT