Whole-slide image (WSI) optimization — toward sustainable digital pathology.
- Convert — WSIs are fragmented. Consolidate them into clean TIFF/OME-TIFF.
- Downsample — WSIs are heavy. Normalize 40x scans to 20x and save storage by ~75%.
--scale 20x - Standardize — Colors are not portable. Bake ICC profiles into the pixels.
--icc-bake - Crop - WSIs are huge. Keep only the annotated regions.
--roi <file.geojson|DIR>
| Input | default | --scale 20x |
--icc-bake |
|---|---|---|---|
| DICOM | 🔵 | 🟢 | 🟢 |
| SVS / TIFF | —² | 🟢 | 🟢 |
| VSI (CellSens)¹ | 🔵 | 🟢 | 🟢 |
🔵 repackaging — compressed tiles are copied straight through, at near-copy speed. No image quality change.
🟢 re-encoding — only where the pixels actually change.
Output is OME-TIFF by default, or SVS-like BigTIFF with --openslide (matches the
source's original compression, and is readable by OpenSlide).
¹ Experimental reader. 8-bit brightfield only.
--scale <number> is not supported for VSI yet.
² Skipped unless combined with --scale, --icc-bake or --roi.
If you'd rather not use the terminal, thinslide-gui does the same thing in a window.
- Download
thinslide-macos-arm64.dmg(macOS) orthinslide-gui-windows-x86_64.exe(Windows) from the latest release. - On macOS, open the .dmg and drag ThinSlide into Applications. On Windows, just run the .exe.
- Open it, then choose a folder of slides and a destination folder.
- Pick what you want, and click Run.
macOS security warning? Since ThinSlide isn't notarized by Apple, macOS may block it on first launch ("cannot be opened because the developer cannot be verified"). Go to System Settings > Privacy & Security, scroll down to the message about ThinSlide, and click Open Anyway. Confirm in the dialog that follows, then launch ThinSlide again.
thinslide <input_dir> <output_dir> [options]Input and output are directories — ThinSlide processes every slide it finds, mixed formats included.
# Convert a folder of DICOM slides into OME-TIFF (uses all CPUs)
thinslide /data/dicoms /data/output
# Same, but write SVS-like BigTIFF instead, OpenSlide-compatible
thinslide /data/dicoms /data/output --openslide
# Normalize everything to 20x
thinslide /data/slides /data/output --scale 20x
# Bake ICC profiles into pixels, output sRGB JPEG
thinslide /data/slides /data/output --icc-bake
# Both at once, in a single pass, tuning quality and threads
thinslide /data/slides /data/output --scale 20x --icc-bake --quality 90 -j 4OME-TIFF inputs keep their original OME-XML metadata through downsampling.
See doc/installation.md for prebuilt binaries and build-from-source instructions.
See doc/advanced.md for full details on --scale and --roi.
See doc/phi-handling.md for details.
ThinSlide's CellSens (.vsi) and MIRAX (.mrxs) readers were developed with reference to, and in part ported from, the following open-source projects:
- Bio-Formats (GPLv2) — CellSens VSI format parsing
- OpenSlide (LGPL-2.1) — MIRAX format parsing
Copyright (C) 2026 Wataru Uegami, MD, PhD
ThinSlide is licensed under the GNU General Public License v2.0 or later (GPL-2.0-or-later).
ThinSlide is provided for research use only. It is not a medical device, has not been cleared or approved by any regulatory authority, and is not intended for clinical diagnosis, treatment, or any patient-care decision. The software is provided "as is", without warranty of any kind, to the extent permitted by applicable law.