iSyntax Converter

Python tool that uses Philips' SDK to write slides in an intermediary raw format.

Requirements

Python 3.6+
Philips iSyntax SDK (https://www.openpathology.philips.com)

The iSyntax SDK must be downloaded separately from Philips and the relevant license agreement agreed to before any conversion can take place.

As of version 0.4.0, which has a Python 3.6+ requirement, the supported iSyntax SDK versions and environments are as follows:

iSyntax SDK 1.2.1 (CentOS 7, Ubuntu 18.04, Windows 10 64-bit)
iSyntax SDK 2.0 (CentOS 8, Ubuntu 18.04, Windows 10 64-bit)

Usage

Basic usage is:

isyntax2raw write_tiles /path/to/input.isyntax /path/to/directory.zarr

Please see isyntax2raw write_tiles --help for detailed information.

Output tile width and height can optionally be specified; default values are detailed in --help.

A directory structure containing the pyramid tiles at all resolutions and macro/label images will be created. The default format is Zarr compliant with https://ngff.openmicroscopy.org/0.4. Additional metadata is written to a JSON file. Be mindful of available disk space, as larger .isyntax files can result in >20 GB of tiles.

Use of the Zarr file type will result in losslessly compressed output. This is the only format currently supported by the downstream raw2ometiff (as of version 0.3.0).

Background color

Any missing tiles are filled with 0 by default, which displays as black. The fill value can be changed using the --fill_color option, which accepts a single integer between 0 and 255 inclusive. Setting --fill_color=255 will cause any missing tiles to display as white.

Performance

This package is highly sensitive to underlying hardware as well as the following configuration options:

--max_workers
--tile_width
--tile_height
--batch_size

On systems with significant I/O bandwidth, particularly SATA or NVMe based storage, we have found sharply diminishing returns with worker counts > 4. There are significant performance gains to be had utilizing larger tile sizes but be mindful of the consequences on the downstream workflow. You may find increasing the batch size on systems with very high single core performance to give modest performance gains.

In general, expect to need to tune the above settings and measure relative performance.