Unbiased Stereology.
AI Precision.
The Optical Fractionator with machine-learning detection — hardware independent, statistically rigorous, and a fraction of the cost of legacy systems.
Stop Overpaying for Cell Counts
Three problems every stereology lab faces — and how ConductStereology solves them.
The Hardware Trap
Legacy stereology locks you into $40,000 motorized stages and proprietary cameras. ConductStereology works on any whole-slide image — zero additional hardware.
The Manual Grind
Marking thousands of cells across hundreds of sampling sites takes weeks of analyst time. AI-assisted detection inside the counting frame cuts that to hours.
The Black-Box AI Problem
Generic AI counters ignore guard zones and counting rules, producing biased estimates reviewers will reject. ConductStereology applies AI within the stereological probe — every detection is validated by you.
Design-Based Stereology Meets Machine Learning
Four capabilities that set ConductStereology apart from legacy systems and manual workflows.
The Optical Fractionator
Systematic random sampling estimates total cell population from a fraction of counting sites. Define your ROI, set the sampling grid, and let the software handle the math.
Implements the method of West, Slomianka & Gundersen (1991) — the gold standard for unbiased cell counting accepted by Nature, Science, and PNAS.
Human-in-the-Loop AI
The AI suggests candidate cells inside the counting frame. You accept or reject each detection — the system learns your criteria while you maintain full control.
Pre-trained models for DAPI (nuclei), NeuN/Nissl (neurons), and Iba1 (microglia). Train custom models on your own staining protocol.
Live Coefficient of Error
CE updates in real-time as you count. Stop when your estimate reaches statistical confidence — no wasted counting beyond what the data requires.
Gundersen CE (m=1) calculated after each sampling site. Most studies achieve CE ≤ 0.05 within 200–400 sites.
True 3D Disector
Scroll through Z-stacks with a virtual optical disector. Only nuclei that come into focus within the counting frame are counted, eliminating lost-cap and overprojection bias.
Guard zones configurable for top and bottom of section thickness. Compatible with confocal and widefield Z-stacks.
Built on the Gold Standard
ConductStereology implements the Optical Fractionator method established by West, Slomianka & Gundersen (1991) — the most widely cited design-based stereology protocol in neuroscience. Every probe follows the same systematic random sampling principles that have been validated across thousands of peer-reviewed publications.
Unbiased sampling grid ensures every location has equal probability of being selected
Inclusion/exclusion lines eliminate edge effects and double-counting
Configurable top/bottom guard zones prevent lost-cap artifacts in thick sections
Live coefficient of error tells you exactly when to stop counting
Cells touching green (inclusion) lines are counted. Cells touching red (exclusion) lines are excluded.
How ConductStereology Compares
Side-by-side with the legacy market leader and manual counting.
| Feature | ConductStereology | Legacy Systems | Manual Counting |
|---|---|---|---|
| Annual license | $1,250–$1,950/yr | $15,000–$40,000 perpetual | “Free” (100+ analyst hours) |
| Hardware required | None — any scanner | Motorized stage + controller | Manual microscope |
| AI-assisted counting | Human-in-the-loop | Not available | Not available |
| CE tracking | Real-time | Post-hoc only | Manual calculation |
| File formats | SVS, NDPI, OME-TIFF | Vendor-specific | N/A |
| Time per section | ~5 min | ~15 min | ~30 min |
| Training time | < 1 hour | Multi-day workshop | Self-taught |
Trusted Across Disciplines
ConductStereology serves any lab that needs unbiased cell population estimates from microscopy data.
- Estimate neuronal density in hippocampus (CA1–CA3), cortex, and subcortical regions
- Compare treatment vs. control groups using unbiased population estimates with CE reporting
- Quantify microglial activation (Iba1+) in neuroinflammation and traumatic brain injury models
- Map dopaminergic neuron loss (TH+) in Parkinson’s disease animal models
- Pre-trained AI models for NeuN, Nissl, and DAPI-stained tissue sections
- Estimate hepatocyte counts and lobular density in drug-induced liver injury studies
- Quantify glomerular density and tubular cell counts in kidney toxicity assessments
- Assess cell proliferation (Ki-67+) in tumor xenograft models for efficacy studies
- Count Purkinje neurons in developmental neurotoxicology dose-response studies
- Track cell proliferation markers (Ki-67, BrdU, EdU) across developmental stages
- Quantify apoptotic cells (TUNEL+, caspase-3+) in embryonic tissue
- Measure neurogenesis rates in subventricular zone and dentate gyrus
- Compare tissue volumes across genotypes using the Cavalieri estimator
From Slide to Statistical Estimate
ConductPrep
Preprocess and stitch whole-slide images or microscopy Z-stacks from any scanner.
ROI Definition
Outline the anatomical region — hippocampus, cortex, liver lobule, or any custom boundary.
Sample & Count
Run the AI-assisted Optical Fractionator on systematically random sampling sites.
Statistical Report
Export total population, volume, density, and CE to Excel or CSV.
Gold Standard Counting. Modern Pricing.
Design-based stereology shouldn’t cost more than the microscope. Every plan includes full access to all probes and AI models.
5-Year
- All stereology probes
- Pre-trained AI models
- Real-time CE tracking
- Excel/CSV export
- Email support
- 5-year price lock
3-Year
- All stereology probes
- Pre-trained AI models
- Real-time CE tracking
- Excel/CSV export
- Priority support
- Training session included
1-Year
- All stereology probes
- Pre-trained AI models
- Real-time CE tracking
- Excel/CSV export
- Priority support
- Cancel anytime
Enterprise & Institutional Licensing — Multi-seat discounts, centralized admin, on-site training, and custom AI models. Contact Sales →
Technical Specifications
- Whole-Slide Images: SVS (Aperio), NDPI (Hamamatsu)
- Microscopy Stacks: OME-TIFF, standard TIFF Z-stacks
- Compatibility: Any microscope or scanner that produces standard formats
- Optical Fractionator: Systematic random sampling for total population estimates
- Cavalieri Estimator: Volume estimation from serial sections
- Optical Disector: 3D counting in Z-stacks with configurable guard zones
- Nuclei (DAPI): General nuclear detection for any tissue type
- Neurons (NeuN/Nissl): Neuronal marker detection in brain tissue
- Microglia (Iba1): Microglial marker detection for neuroinflammation studies
- Custom Training: Train models on your own staining protocol
- Statistics: Total population, volume, density, CE (Gundersen m=1)
- Formats: Excel (.xlsx), CSV, JSON
- Reporting: Per-section and aggregate results with sampling parameters
Frequently Asked Questions
Do I need a motorized microscope?
No. ConductStereology works with pre-scanned whole-slide images. Use any slide scanner — Aperio, Hamamatsu, Leica, or any system that produces SVS, NDPI, or OME-TIFF files. Your existing scanned images work immediately.
Is the AI validated for peer-reviewed publication?
Yes. The human-in-the-loop approach means every AI detection is validated by the researcher before being counted. The statistical framework follows the Optical Fractionator method (West et al., 1991), the accepted standard in quantitative neuroscience journals including Nature Neuroscience, Journal of Neuroscience, and Brain Research.
How does the Coefficient of Error work?
CE is calculated using the Gundersen formula (m=1) and updates live after each sampling site. When CE drops below your target threshold (typically 0.05), you have sufficient statistical confidence to stop counting — no wasted effort beyond what the data requires.
Can I use images I’ve already scanned?
Absolutely. Import any existing whole-slide images or Z-stacks. ConductStereology reads standard formats (SVS, NDPI, OME-TIFF) from any scanner vendor.
What AI models are included?
Pre-trained models for DAPI (nuclear staining), NeuN/Nissl (neuronal markers), and Iba1 (microglial markers). You can also train custom models on your specific staining protocol directly within the platform.
How does this compare to Stereo Investigator?
ConductStereology provides the same design-based stereological methods at a fraction of the cost, without requiring proprietary hardware. The key addition is AI-assisted cell detection inside the probe, which accelerates counting while maintaining full researcher control over every mark.
Can multiple lab members share a license?
Each subscription is per-seat. For multi-user licensing, our Enterprise plan offers volume discounts, shared project workspaces, and centralized administration for institutional deployments.
What support and training is included?
All plans include documentation, video tutorials, and email support. 3-Year and 1-Year plans include priority support with direct access to our stereology specialists. The 3-Year plan includes a live training session.
Part of the ConductScience Pipeline
ConductStereology integrates with ConductPrep for preprocessing, providing a seamless path from raw images to unbiased quantification.
Microscopes
ConductPrep
ConductStereology
PublicationsReady to Modernize
Your Stereology?
Request a demo and see how ConductStereology can accelerate your lab’s cell counting workflow.
