FluorocellAI — Imaging & Cell Analysis

AI-Powered Cell Segmentation.
Built Into the Tools You Already Use.

Deep learning cell segmentation, nucleus detection, and automated quantification — running natively inside ImageJ and Fiji. Priced for the independent investigator. Accurate enough for the core facility.

>104
Cells segmented / hour / node
>95%
Mean F1 on held-out benchmarks
2.8M+
Annotated cells in training corpus
8
Simultaneous fluorescence channels
12
Cell lines in training corpus

The annotation bottleneck in quantitative cell biology is not a data problem — it is a labour problem. High-content screening instruments generate tens of thousands of fields per experiment. The cells in those images do not label themselves. Manual correction at that scale is not a method; it is a funding liability. FluorocellAI was built to eliminate that liability without asking the investigator to become a machine learning engineer.

FluorocellAI — Scientific Rationale
Analysis Pipeline
01
Image Ingestion

TIFF, CZI, LIF, ND2, ICS/IDS. Direct OMERO server connection. No format conversion at any stage.

02
Preprocessing

Illumination correction, background subtraction, flat-field normalization, multi-channel registration.

03
AI Inference

Transformer encoder + CNN decoder ensemble. Nuclear detection, boundary delineation, cell-cycle phase classification.

04
QC & Interactive Review

Per-field confidence scoring, outlier flagging, manual correction via native ImageJ ROI Manager.

05
Export

CSV, HDF5, ZARR, label TIFF, ROI Manager. Direct to CellProfiler, R, Python, or custom pipelines.

ImageJ Update Site

Installs in three clicks via the Fiji update site mechanism. No separate application. No workflow disruption.

Scientific Context

Two decades of throughput advancement in microscopy acquisition — with the labeling workflow unchanged.

High-content fluorescence microscopy platforms — Operetta, CQ1, IN Cell Analyzer — routinely generate datasets of tens of thousands of fields per experiment, each containing hundreds to thousands of individual cells. The acquisition side of this problem has been solved. The downstream side has not. Identifying individual cells within those images remains the primary constraint on throughput in quantitative cell biology.

Classical segmentation approaches fail systematically on the conditions most relevant to high-content work. Otsu and multi-level thresholding cannot resolve boundaries between adjacent cells in confluent monolayers. Watershed applied to nuclear DAPI staining produces systematic errors in S and G2 phase cells where nuclear morphology departs from the ellipsoidal model the algorithm assumes. These are not edge cases — they are the normal operating conditions of the assays that generate the most biologically important data.

FluorocellAI replaces this bottleneck with a purpose-trained deep learning inference pipeline. The underlying model was trained on a curated dataset of >2.8 million manually annotated cells across twelve cell lines, six imaging platforms, and a broad range of staining conditions — including confluent monolayers, three-dimensional organoids, and primary cultures with irregular morphologies that generalist models handle poorly.

The system integrates within ImageJ and Fiji — the environments the cell biology community has standardized around — preserving accumulated analytical infrastructure: macros, measurement pipelines, plugin ecosystems. FluorocellAI adds capability without extracting ownership of the workflow. The investigator retains full control.

Architecture
Transformer encoder + CNN decoder ensemble
Input Formats
TIFF, CZI, LIF, ND2, ICS/IDS
GPU Acceleration
CUDA 11+; cloud inference optional
Min. Configuration
8-core, 16 GB RAM, Java 11+, Fiji 2.14+
Core Capabilities

Every segmentation challenge in quantitative cell biology — addressed within a single ImageJ plugin.

Nuclear Segmentation
DAPI / Hoechst

Precise delineation of individual nuclei in densely packed monolayers and three-dimensional organoid sections. Handles touching and overlapping nuclei through learned instance separation — not post-hoc watershed. Validated across HeLa, U2OS, A549, MCF-7, primary hepatocytes, iPSC-derived neurons, and murine embryonic fibroblasts.

Whole-Cell Boundary
Cytoplasmic membrane

Cytoplasmic membrane segmentation for cells expressing membrane-targeted fluorescent proteins or stained with cytoplasmic dyes. Accurately resolves boundaries in confluent monolayers where adjacent membranes are in direct contact. Produces per-cell area, perimeter, eccentricity, solidity, and texture features compatible with CellProfiler pipelines.

Multi-Channel Analysis
Up to 8 channels

Spectral co-localization and intensity quantification across up to eight simultaneous fluorescence channels. Automated per-cell Pearson's r and Manders' M1/M2 coefficients. Population-heterogeneity analysis at the single-cell level — not the image level — which bulk statistics systematically obscure.

Cell Cycle Classification
G1 / S / G2 / M

Automated phase assignment using integrated nuclear DAPI intensity combined with EdU or BrdU incorporation signals where available. In the absence of S-phase markers, morphology and integrated intensity distributions construct the G1/G2 diploid-tetraploid ratio. Phase assignments exported as per-cell categorical variables for population-level analysis.

HCS Batch Execution
96- and 384-well plates

Plate-level batch execution for high-content screening datasets. Automated per-well QC: focus scoring, cell density assessment, saturation flagging, edge-effect detection. Aggregated phenotype statistics export in formats compatible with KNIME, CellProfiler Analyst, and R/Bioconductor. Z-score and SSMD calculations for hit identification pipelines.

3D Volumetric Segmentation
Confocal / LSFM / 2-photon

Extension of the segmentation pipeline to z-stack volumes acquired by confocal, spinning-disk, light-sheet, or two-photon modalities. Produces volumetric cell meshes with 3D morphometric features: volume, surface area, sphericity, principal axes, inter-cell contact area. Compatible with Imaris, Napari, and the Fiji 3D Viewer.

Live-Cell Tracking
Timelapse fluorescence

Frame-to-frame tracking using nearest-neighbor with learned appearance features for disambiguation at track crossings. Produces cell lineage trees, migration trajectories, and per-cell temporal feature vectors. Validated at frame intervals from 2 minutes to 2 hours across widefield and confocal timelapse protocols.

Macro & Script API
ImageJ / Python / REST

Full programmatic access via the ImageJ Macro Language, BeanShell, Groovy, and Python through the SciJava framework. Enables embedding within custom pipelines, unattended HPC batch processing, and LIMS integration. REST API available for cloud inference deployments.

Benchmark Performance

Validated against the published state of the art on five public datasets.

All F1 scores at IoU threshold 0.5. Cellpose 2.0 and StarDist results reproduced from published benchmarks using default parameters. FluorocellAI results use the pretrained general model without dataset-specific fine-tuning.

Comparison tools
FluorocellAI
BBBC039 — Nuclei
U2OS DAPI, 96-well
Cellpose 2.0
0.871
StarDist
0.893
FluorocellAI
0.941
BBBC006 — z-stack
U2OS, multi-plane DAPI
Cellpose 2.0
0.834
StarDist
0.861
FluorocellAI
0.908
DSB 2018
Mixed types, multi-modality
Cellpose 2.0
0.856
StarDist
0.879
FluorocellAI
0.924
CTC HistoSeg
HeLa, phase contrast
Cellpose 2.0
0.791
StarDist
0.803
FluorocellAI
0.867
3D Organoid (internal)
Intestinal organoids, confocal z
Cellpose 2.0
0.762
StarDist
0.781
FluorocellAI
0.851
Integration Ecosystem

Your analytical infrastructure took years to build. FluorocellAI augments it.

ImageJ / Fiji

The primary integration. FluorocellAI lives inside Fiji.

FluorocellAI installs as a standard Fiji plugin via the update site mechanism — the same mechanism used by every other plugin in the Fiji ecosystem. All segmentation outputs are written as native ImageJ ROI objects into the ROI Manager, making them immediately accessible to any downstream ImageJ operation without format conversion, data export, or workflow interruption.

The full ImageJ Macro Language API is exposed, enabling FluorocellAI inference to be scripted into automated pipelines running unattended on large image datasets overnight on institutional compute infrastructure.

OMERO

Server-side batch segmentation of managed repositories.

Connects directly to the OMERO.server API. Segmentation results stored back as ROI and measurement table annotations, preserving full data provenance within your image management system. Compatible with OMERO 5.6+.

CellProfiler

Superior segmentation inputs into existing measurement pipelines.

Export 16-bit label TIFF images as IdentifyPrimaryObjects input. Existing CellProfiler pipelines require no modification to downstream steps — only the segmentation step improves.

Python / Napari / R

Native arrays for scripting environments.

ZARR and HDF5 label arrays readable by NumPy and scikit-image. Napari-compatible label layers for interactive correction. A companion yashara.r package parses FluorocellAI HDF5 output into tidy data frames.

Frequently Asked Questions

Technical and operational questions answered.

Does FluorocellAI require GPU hardware?+
FluorocellAI runs on standard CPU hardware with no additional requirements beyond a modern multi-core processor and at least 16 GB of RAM. GPU acceleration via CUDA is supported and recommended for throughput-intensive applications — a consumer-grade NVIDIA GPU (RTX 3060 or equivalent) reduces per-image inference time by approximately 6–8×. Cloud-based GPU inference is available as an option for users without local GPU hardware.
Does the pretrained model require retraining for my cell type?+
The pretrained FluorocellAI general model performs well on the majority of standard cell lines and imaging conditions without any retraining. For unusual cell morphologies, sparse cultures, or non-standard staining conditions, a fine-tuning workflow is provided that requires as few as 50 manually annotated cells from your specific imaging context.
What version of Fiji is required?+
FluorocellAI requires Fiji version 2.14.0 or later, running on Java 11 or later. The standard Fiji distribution includes Java 11 as a bundled component, so no separate Java installation is typically required. Plain ImageJ (non-Fiji) is not supported due to dependency requirements not available in the base distribution.
How does group licensing work?+
Laboratory group licenses cover all members of a defined research group under a single subscription, with per-seat pricing applied. Group membership is managed through the Yashara account portal and can be updated as laboratory personnel changes. Core facility site licenses are available covering unlimited users within the licensed institution.
Is my image data processed on Yashara's servers?+
By default, FluorocellAI performs all inference locally on your own hardware. Image data does not leave your institution's network unless you explicitly opt into the cloud inference feature. The local inference pathway requires no network connection after initial installation and license activation.
Pricing

Accessible at every scale — from the single-investigator laboratory to the institutional core facility.

Individual Investigator
Contact Us
per seat / year
  • Single-user license
  • Full segmentation feature set
  • Local CPU and GPU inference
  • Email support (48h SLA)
Get started →
Core Facility / Institution
Custom
Site licensing available
  • Unlimited-seat site license
  • OMERO server-side integration
  • Dedicated technical contact
  • Macro API and REST API access
  • White-label deployment option
Contact us →
Contact the FluorocellAI Team

Reach us directly.

Tell us your cell type, imaging modality, and what you are trying to solve. We respond within two business days with a substantive reply — not a boilerplate acknowledgement.

Direct Line

For urgent or highly technical questions, write directly to the FluorocellAI team. This address is monitored by the scientists who built the segmentation pipeline, not a general support queue.

fluorocell@yashara.us
Evaluation Access

We provide structured 30-day evaluation access for qualified research teams. To make the evaluation useful rather than generic, we configure it to your specific cell type and imaging modality where possible. Tell us your context in the message field.

Demonstrations

Product demonstrations are technical conversations, not sales presentations. If you tell us your cell line and imaging setup in advance, we come prepared with relevant segmentation examples — not a deck designed for a general audience.

Response Time

We respond to all substantive inquiries within two business days. Replies come from a scientist, not a CRM sequence.