CellCount.

Reproducible cell counts, through your own microscope. Trypan blue viability in seconds, whether the image comes from a chamber under the microscope or from the counter the lab already owns, with every count returned on its image, with the metrics behind it.

Open cellcount.ai Try the live demo

Built on the hardware the lab already owns.

A camera on the scope

A plug-and-play camera on the microscope port, counting straight into CellCount on the laptop at the bench. The recommended setup.

A phone at the eyepiece

Hold a modern phone to the eyepiece and count from the photo. No adapter, no hardware at all.

Images you already have

Drop in files from any microscope, or a whole export folder from an automated counter, and the same algorithm runs on them as on a chamber under a microscope.

An inverted microscope with a Scope Cam on its camera port, and beside it a touchscreen PC showing CellCount with the counted grid and the concentration in cells per mL — A Scope Cam on the microscope and CellCount on a touchscreen PC beside it, at ShareLab.

See it in use.

Short videos of real counts in CellCount, from the photo of the chamber to live and dead cells, cells per mL and viability.

CellCount showing a counted hemocytometer image with live cells in green and dead in red, and 90.4% viability on the result card, beside a presenter in a circle.

0:40

Count cells from one hemocytometer photo

One photo of the chamber, and CellCount finds the grid, sets the scale and counts the cells. One tap adds trypan blue viability.

CellCount on a phone screen showing a counted chamber image with 90.4% viability, beside a presenter in a circle.

0:42

Count cells on your phone

CellCount in the phone’s browser, with a photo through the eyepiece: the grid, every cell outlined, and trypan blue viability.

CellCount counting the whole image of a chamber photo, with 98.3% viability on the result card, beside a presenter in a circle.

1:44

Full image mode: count every cell in the frame

When the whole grid does not fit, count the whole image: the grid still sets the scale, and the counted volume grows.

One upload, the whole answer, and the record.

The count, on the image

Every detected cell marked on the lab’s own frame, the grid it measured, the concentration in cells per mL from that grid and the dilution, the viability, and a cell-size histogram with the gate the lab set.

One protocol for the whole lab

Dilution, which squares, sensitivity and size gate set once and shared by the team, so counts agree between people, instruments and days.

A record someone signed

Each sample is dated, accepted by a named person and sealed with a hash. It leaves as a results document, or as an export bundle of the images with their overlays and results, for the lab’s own archive.

Tested on frames it never saw.

Measured against the instrument labs already trust: sixteen images straight from a leading automated cell counter, which CellCount had never seen. The method, the plots and the sources are on the validation page.

0.91Lin’s concordance on cell concentration with a leading automated cell counter, across sixteen images

±0.29million cells/mL, the 95% limits of agreement, with a mean difference of 0.002

95%the same live-or-dead call on the 6,765 cells both readers found

0images from that instrument in CellCount’s training data

Read the validation Questions labs ask Book a demo on your samples

Free to read and use, no account.

Guides and a calculator for counting cells by hand, on cellcount.ai.