The OptiFlow platform

One platform.
Many pipelines, end to end.

OptiFlow is a modular, imaging modality-agnostic analytics platform. Build domain-aware workflows visually, run them from a single image to millions, and turn raw images into measurements, charts, and finished reports — no code, 100% local.

Imaging modality-agnostic No-code workflow builder Domain-aware AI Tamper-evident audit trail Air-gapped by design
The architecture

From raw image
to decision you can defend.

One coherent path runs through every deployment. Point OptiFlow at your images, compose a workflow visually, run it through the modules your work demands, and export results — each one carrying the exact record of how it was made.

Step 01

Any image source

Files, whole folders, or live feeds — any imaging modality. Pixel calibration is read straight from your metadata.

Step 02

Visual workflow

Drag in operations, branch and combine, and preview every step on your own images.

Step 03

Domain-aware modules

Restore, detect, classify, and quantify with models built for your field — not generic ones repurposed.

Step 04

Decisions & exports

Annotated images, measurements, charts, and formatted reports — exported to TIFF, PNG, Excel, and PDF.

Provenance is embedded at every step. Every output records the exact operations, parameters, calibration, and operator that produced it — written into the file itself, ready for audit or replication.

The workflow builder

Built by one specialist.
Run by your whole team.

Compose serious pipelines without a programming team. Branch one cleaned image into several analysis paths, and preview any step on your own data before you commit. When it works, save it — and anyone can run it.

Preview any step Branch & combine Inspect intermediates Batch at scale Save & share workflows Measure in real units
The module library

Preview everything.
License what you need.

Every workflow is assembled from a deep library of domain-aware operations — from signal recovery to classification to reporting. Try any of them on your own data, then license only what your work demands.

Restore & enhance

Recover signal. Correct the acquisition.

Denoising that is classical, AI-guided, and hybrid — NLM, bilateral, wavelet, total-variation, plus BM3D for heavy noise and Perona–Malik anisotropic diffusion. Destriping for FIB-SEM curtaining. Illumination, contrast and CLAHE. Frequency-domain filtering with FFT notch and Butterworth band-pass.

Expanded

Detect, segment & measure

Find structures. Quantify them.

Otsu, multi-Otsu and adaptive thresholding; morphology and connected-component labelling; edge detection (Canny, Sobel, Scharr); and ridge filters (Frangi, Sato, Meijering) for vessels, filaments and fibres — or traces, cracks and grain boundaries. Calibrated distance, area, and interactive line-intensity profiling.

New

Classify & grade

Decisions trained on your data.

Supervised deep learning for cell type, phenotype, disease grade, or defect class — trained on your own annotated images, so the model learns your problem rather than approximating it.

Anomaly & defect detection

Catch the known and the unknown.

Supervised detection against a defect library for failure modes you already know, plus unsupervised anomaly detection that flags novel defects with no labels required.

Label-free imaging CytoByte

Nuclear morphology, without a stain.

In-silico staining turns DIC images into a virtual nuclear stain in real time — visualising structure without a single chemical stain. In production at ETH Zürich.

Quantify, profile & report

From pixels to a finished report.

Per-object and population statistics, texture analysis, interactive line-intensity profiles, and built-in plotting — histogram, scatter, box-plot, violin, CDF. Export the whole batch to Excel automatically, or compile results into formatted PDF reports with a structure tailored to your team.

New

Plus the building blocks every pipeline leans on

Channel and colour operations, lookup tables, geometry and alignment, type conversion, and pixel-wise image maths and compositing — the connective tissue that lets any workflow fit together exactly as your work requires.

Modular by design — add modules as your use cases grow, instrument by instrument and team by team. Explore the engines built for biomedical, semiconductor, or your own domain.

Integrity by design

Defensible results.
Private by default.

Integrity isn't a premium tier or a bolt-on. It is the architecture. Every result is reproducible, every deployment stays on your own machines, and the record can be trusted.

Embedded provenance

Every result carries its full recipe.

The complete pipeline — every operation, parameter, calibration and operator — is written into the image file itself, encrypted and readable only inside OptiFlow. Re-process a file and the new run appends to the chain; where it matters, the record can be locked to a password only your team holds.

Air-gapped & 100% local

No image ever leaves the machine.

Processing reads and writes your local filesystem directly. No image data crosses the network, no cloud account, no exceptions — it runs fully offline, inside your own environment.

Tamper-evident & licence-bound

Trust the record. Control the deployment.

An encrypted, rollback-resistant timestamp guards against clock tampering, and cryptographically signed, hardware-bound licences keep execution on the machines you authorise.

See it: a workflow, loaded straight from a result.

Every result already contains the workflow that made it. So when you want to revisit your own analysis, hand it to a colleague, satisfy a reviewer, or re-run it across a folder, the full method is right there — nothing to document, nothing to rebuild. And it stays yours: the record is encrypted, readable only inside OptiFlow, and accessible only when you choose.

Preparing live demo…
Deployment & integration

Fits your environment.
Runs on day one.

A self-contained desktop application that sits beside the tools you already use — no servers to provision, no workflow to rebuild from scratch.

Desktop-native

Runs as a self-contained Windows application. No cloud account, no server to stand up, nothing to provision.

Works with your stack

Sits alongside your existing instruments and file stores. Point it at a folder and go — nothing to migrate.

Open, portable outputs

Exports to TIFF, PNG and Excel. Workflows save as portable files you can share, version, or hand to a colleague.

Reads your calibration

Picks up pixel size from ImageJ, OME-TIFF and TIFF metadata, so measurements land in real units — nm or µm — automatically.

How we work

Start small.
Scale with confidence.

No rip-and-replace, no leap of faith. Prove the value on your own data first, then grow at your own pace.

Explore Deploy Expand Embed
Preview before you buy 60-day module guarantee Priority support, included Pay only for what you use
In production at ETH Zürich Ongoing trials at NTU 20+ peer-reviewed publications Featured by CNA & The Straits Times

See it on your images.

Bring us a problem and your own data. We'll show you exactly what OptiFlow does with them.