olympus-browser-qt

Reusable PyQt6 dialog and viewer for browsing Olympus / Evident CellSens .vsi datasets.

This package follows the same public shape as the Leica, Zeiss, and Nikon Qt browsers, but uses a native Python VSI/ETS reader. It does not use Bio-Formats or Java.

Install

pip install -e .

For this checkout, run_viewer.cmd uses the local deconvolve conda environment when it is available.

Current Status

• Scans folders for .vsi files and exposes Olympus image contexts.
• Reads TIFF preview/SIS metadata from the .vsi container with tifffile.
• Discovers companion ETS files in _<vsi stem>_/stack*/frame_*.ets.
• Natively parses the Olympus SIS/ETS chunk table for raw uncompressed ETS files.
• Reads real uint8, uint16, integer, and float raw ETS planes and regions.
• Generates real previews from ETS planes, with TIFF preview fallback.

Compressed ETS variants such as JPEG, JPEG2000, PNG, BMP, and lossless JPEG are recognized by compression code but are not decoded yet.

Single Select

from PyQt6.QtWidgets import QApplication
from olympus_browser_qt import OlympusBrowserDialog

app = QApplication([])
ctx = OlympusBrowserDialog.select_image_context(roots=[r"D:\data"])
if ctx is not None:
    print(ctx.name, ctx.container_path, ctx.size_x, ctx.size_y)

Direct Pixel Reads

from olympus_browser_qt import OlympusGateway

node = OlympusGateway().container_node(r"D:\data\sample.vsi")
ctx = node.children[0].context

plane = ctx.open().read_plane(z=0, c=0)
stack = ctx.open().read_stack(c=0)
arr = ctx.open().read_array()

The public handle returns NumPy arrays. For the sample raw ETS dataset, planes are real uint16 arrays read from the external .ets file.

CLI

olympus_browser D:\data
olympus_browser D:\data\sample.vsi --multi
olympus_viewer
run_viewer.cmd