Skip to content
Aiosyn

Neuron loss and neuroinflammation,measured per brain region

Two questions that recur across most brain studies, answered as numbers per region instead of a semiquantitative score. Aiosyn's AI delineates the anatomical regions on each stained section, then detects individual NeuN-positive neurons or Iba1-positive microglia within them, across the whole section under one fixed method.

Discuss your studyBiopharma overview
  • Continuous numbers, no semiquantitative score
  • Whole section measured, no sampled fields
  • Same section, same density, every run
NeuN sectionDetected neurons
Drag the handle. Left: the stained hippocampus. Right: every neuron the AI detected, one mark each.

Cortex, hippocampus, thalamus and hypothalamus, delineated first

Every read-out that follows is reported inside those boundaries, so a density from one animal is comparable with the same density from the next, and sections cut at slightly different levels still report against the same compartments.

Stained sectionDelineated regions
Left: the stained section. Right: the compartments every count is reported in, delineated across the whole tissue.
  • Cortex
  • Hippocampus
  • Thalamus
  • Hypothalamus

Has neuron density changed, and where?

Every NeuN-positive neuron on the section is detected and attributed to the region it sits in, so neuron loss is reported as a density per region instead of an impression of a thinner band. The section in this figure carries more than 30,000 neurons, which is why this read-out is infeasible to produce by hand.

A whole brain section with every detected neuron marked in green, inside the delineated regions.
Every neuron on the section, detected and attributed to its region. More than 30,000 of them on this one slide.

Is there neuroinflammation, and how is it distributed?

Iba1-positive cells are detected individually across the whole section, yielding microglial density per anatomical region rather than scores from manually selected fields. That makes a region-specific change visible and comparable across regions, not averaged into one section-wide number.

Choose a view of the microglia detection
An Iba1-stained field with each microglial cell the AI detected marked individually.
A whole brain section with detected Iba1-positive microglia marked across the tissue, inside the delineated regions.
One mark per microglial cell up close, and the same detection counted inside every region across the whole section.

Cortical thickness, along the full ribbon

The same delineations that carry cell densities also carry structural measurements. Cortical thickness is the one we are asked for most, measured from the segmented cortex along the full ribbon rather than at a few points chosen by hand.

Which structural read-outs your study gets is settled when we scope it, against your sections and the endpoint you need. Tell us what the study turns on and we will tell you what we can measure for it.

Ask about a structural read-out
A brain section with cortical thickness measured along the cortical ribbon and rendered as a color map. Each point on the inner cortical boundary represents a measurement point, with color encoding thickness from thin (blue) to thick (red).
Cortical thickness measured along the cortical ribbon, rendered as a color map. Each point on the inner boundary is a measurement, colored thin (blue) to thick (red).

Judge it on your own sections

The region models and the cell detectors were developed with pathologists and preclinical researchers who read brain tissue, on annotated sections.

The kidney and atherosclerosis read-outs each have a validation behind them. For brain, that comparison is not published, so the way to weigh these read-outs is your own tissue: send a handful of sections and you get the densities back with every detection visible on the image it came from, next to the counting you do today.

Send us a few sections

From slide to per-region read-out

Four steps. Your scanning workflow does not change, and no manual counting happens in between.

  1. 1

    Whole-slide scan

    Stained brain sections, digitized in the formats you already scan.

  2. 2

    Regions mapped

    Anatomical compartments are delineated across the full section.

  3. 3

    Cells detected

    Neurons and microglia are detected individually and assigned to a region.

  4. 4

    Read-outs returned

    Per-region densities, per animal and per experimental group, with the detections they were counted from.

A marker that is not on this page

Another antibody, another region set or a read-out specific to your model is built on your own images as custom AI development.

Co-developed with TNO Health & Work

Through our partnership, we co-develop these models together with TNO. Their preclinical and pharmaceutical research expertise sits behind the study design and the tissue, our AI behind the read-outs, and the models are applied within TNO's preclinical studies with more study areas in development.

Continue with another tissue

Biopharma overview

Bring us a brain cohort

Tell us the marker and the regions you need quantified. We will return the read-outs on a few of your own sections so you can judge them against your current counting.