Funded by EU Horizon Europe · OSCARS Initiative

Similarity-Preserving Content Codes
for Bioimaging Data

BIOCODES implements the ISO 24138 International Standard Content Code for bioimaging data — enabling integrity verification, deduplication, and cross-platform provenance to make your data AI-ready and FAIR-compliant.

Get Started Browse Tools Forum

Apache 2.0 · Open Source ISO 24138:2024 FAIR Principles

Enhancing AI-Readiness of Bioimaging Data
with Content-Based Identifiers

Challenge

Growing volume of data: Bio(imaging) data exist at different states — raw, repository, publications — with no shared identity across them.
No Audit Trail: Challenges to verify data integrity or detect manipulation after the fact.
Lost Provenance: Published figures are disconnected from raw data and the processing steps that produced them.

Solution

International Standard Content Code (ISCC ISO 24138). Open and open-source. Interoperable content identification & fingerprinting system. Computed directly from the asset itself — can never be removed or decoupled from the data.
Anyone can compute ISCCs from available data — independently and without any central authority. Use the ISCC Generator at iscc.io/resources. Learn more
Generating, signing, and timestamping ISCCs creates persistent identifiers to securely reference and link data in repositories with images in papers.

Scientific Impact

Cryptographic figure data verification via the ISCC audit trail.
Improved data integrity, transparency, and reusability.
Enhanced AI-readiness of bioimaging datasets for trusted AI applications.

ISO 24138

International Standard Content Code

A standardised (ISO 24138) multi-component fingerprint for various media types and file formats. Computed from the asset itself — it can never be removed or decoupled from the data.

Semantic level

Detects conceptually related content

Syntactic level

Detects near-duplicate and structurally similar content

Data level

Detects exact copies via cryptographic hash

ISCC Homepage Specification ISO 24138

ISCC-ID

A persistent identifier derived from ISCC content codes. ISCC-IDs link raw data, processed derivatives, and published figures into a single auditable chain of provenance.

Fingerprinting

Helps find metadata even when filenames or paths have changed.

Digital signing

Proves authenticity of the content and its originator.

Timestamping

Demonstrates when content was created or registered.

Secure linking

Enables provenance verification across repositories, publications, and analysis pipelines.

Capabilities

Built for scientific data & bioimaging

Addressing the specific challenges of large-scale imaging data in modern research environments.

ISO 24138:2024 Compliant

Implements the official international standard, ensuring global interoperability across institutions, repositories, and tools.

High Performance

Rust-based engine processes data at 1+ GB/s — up to 184× faster than the pure Python reference implementation, and faster than SHA-256.

Format Agnostic

Works with OME-TIFF, OME-Zarr, CZI, ND2, LIF, DICOM, HDF5 and virtually any binary scientific data format.

FAIR Principles

Supports Findable, Accessible, Interoperable, and Reusable data principles as mandated by EOSC and European funding bodies.

AI-Ready Data

Persistent content-based identifiers survive format conversions and enable reliable provenance tracking for AI training datasets.

Platform Integration

Native plugins for OMERO and Galaxy, with Napari, CellProfiler, and ImageJ integrations in progress.

Open Source Tools

The BIOCODES toolkit

Three complementary tools covering the full bioimaging identification workflow, all Apache 2.0 licensed.

iscc-sum

Stable v0.1

High-performance ISCC Data-Code and Instance-Code generation. Single-pass processing with a Rust core and Python bindings — a drop-in replacement for md5sum and sha256sum in scientific pipelines. Faster than SHA-256 at any data size.

pip install iscc-sum

Platforms: Linux macOS Windows

Formats: Zarr HDF5 OME-TIFF NGFF

Docs Quickstart GitHub PyPI

Rust + Python CLI Apache 2.0

iscc-bio

Beta

ISCC processing for multi-dimensional bioimage data. Implements the IMAGEWALK specification — deterministic Z→C→T plane traversal for format-agnostic, reproducible content hashing of microscopy volumes.

Platforms: Linux macOS Windows

Formats: OME-TIFF OME-Zarr CZI ND2 LIF DICOM HDF5

GitHub PyPI

Python OME-TIFF OME-Zarr CZI / ND2 / LIF Apache 2.0

omero-iscc

Alpha

OMERO server integration plugin. Automatically generates and stores ISCC identifiers for images imported into OMERO, enabling facility-level deduplication and FAIR-compliant provenance tracking.

Platforms: Linux macOS

GitHub

OMERO Plugin Python Server Apache 2.0

Community & Integrations

Built for the open bioimaging ecosystem

BIOCODES integrates with the tools researchers already use — no new infrastructure required.

OMERO

Available

Server plugin for automatic ISCC generation on image import. Enables facility-level deduplication and provenance tracking within OMERO's image management infrastructure.

Automatic ISCC generation on import
Facility-level deduplication
FAIR-compliant metadata annotation

omero-iscc

Galaxy

Available

Galaxy tools for ISCC generation, near-duplicate detection, and content verification within reproducible Galaxy workflows. Part of the BMCV galaxy-image-analysis tool suite.