CellCount is a web app with a free plan that counts cells in hemocytometer images. Photograph the chamber with a phone at the eyepiece or a microscope camera; CellCount finds the grid by itself (if the rulings cannot be found, you can place the crossings by hand), counts live and dead cells by trypan blue exclusion, and gives viability and cells per mL with the dilution you set. It also counts the images an automated cell counter exports, and every detected cell comes back marked on your own image.
No. CellCount finds the hemocytometer grid in your image by itself. The rulings give it the scale and the square to count in, one border rule decides the cells on the lines, and the volume and the cells per mL are worked out from there, with the dilution you set. There are no regions to draw and nothing to calibrate first, so a phone photo at the eyepiece is enough to start. It reads the 1 mm squares and the 0.2 mm medium squares of the centre, so a close-up inside the centre square gets its scale too. If the rulings cannot be found, in a blurred photo for example, you can place the crossings by hand.
Yes. The Free plan counts 180 images a month, every month, shared by up to 3 people on a team, with no card and no end date. Sign up and count today with a phone at the eyepiece or a camera on the microscope. On the Free plan, counted images may be used to improve the service, including our models. Labs that need more get a plan shaped around them.
On held-out frames the shipped model had never seen, its live-or-dead call agreed with an expert annotator on 98.7% of the 2,682 cells both found. Against a leading automated cell counter, on 16 images CellCount had never seen from that instrument, concordance on cell concentration was 0.91 (Lin’s coefficient) and the live-or-dead call agreed on 95% of the 6,765 cells both readers found. The method, the data and every number are kept with the study and are available on request.
A web browser and an image. There are three ways in: a plug-and-play camera on the microscope port counting straight into CellCount on the laptop at the bench, a phone held to the eyepiece, or image files you already have, including a whole folder at once. Nothing to stock and nothing to reorder.
Trypan blue viability counts on a hemocytometer, for immortal cell lines and primary cells such as PBMCs. The quickest way to find out whether it works on your cells is to drop in a few of your own frames and look at the marked image, or to book a demo on your own samples.
Yes. Yeast cells are small, so use a microscope camera at its full resolution, trypan blue, and the High sensitivity setting, which is the preset made for small cells. You get the count, the viability and the cell-size histogram from the same image, and one counting protocol shared across the lab or the brewery. Ask us about other stains on a call.
Does it work with the automated counter my lab already has?
Yes, when the counter can export its raw images. Drop the export folder onto the counting page and CellCount runs the same algorithm on those frames as on a chamber under a microscope, so a facility running both stops arguing about whose number is right.
What about PBMCs, blood-derived samples and clumps?
CellCount is a general-purpose cell detector with viability. It does not classify cell types, so it will not remove red blood cells or debris on its own; the cell-size gate excludes what is too small or too large for your cells, the sensitivity setting decides how faint a detection may be, and every call is shown on your image so you can correct a clump or a wrong mark yourself. The count is not final until you accept it.
Yes. Hold a modern phone to the eyepiece, focus on the chamber rulings, and photograph the chamber with one full 1 mm square and all four of its corners in the frame. CellCount measures the grid in the photo and counts inside it; if the rulings cannot be found, you place the crossings by hand. No adapter and no hardware.
Cells per mL is the average number of cells in one 1 mm square multiplied by 10,000, because a square over the chamber’s 0.1 mm depth holds 0.1 microlitre, and then by the dilution factor; one part sample mixed with one part trypan blue is a factor of two. Viability is live divided by live plus dead. CellCount does the same arithmetic from the grid it measures in your image and the dilution in your protocol.
No. Every cell the model finds is marked on your image. You can add a missed cell, delete a wrong one, flip live to dead, or re-run at another sensitivity, and nothing is final until the person who knows the sample approves the count.
What do I get out at the end?
The count with the image and the metrics behind it, one counting protocol shared by the whole lab so everyone counts the same way, and a report you can export when you need it.
What does one upload give me?
The cell count, the concentration in cells per mL from the measured grid and your dilution, the viability, and a cell-size histogram with the gate you set, all on the same image, in the record and in the results document.
Yes. A protocol holds the dilution, which squares are counted, the sensitivity and the cell-size gate. Anyone on the team can create one, protocols are versioned, and every count in the lab follows the protocol it was made under, whether the image came from a chamber or from your counter’s export.
What is the record of a count?
Each sample is dated and accepted by a named person, and the record is sealed with a SHA-256 hash that shows it has not changed since. The results document you download names who accepted it and when, and carries the images with their overlays, the metrics and the statistical analysis.
Yes. Every sample can be downloaded as an export bundle: the original images, the images with the detected cells drawn on them, the results as CSV, and a manifest that ties them together, so the count can be reviewed later without the app.
Is there an API?
Yes, with a Custom plan. Send a JPEG, PNG or TIFF from your own systems, such as a LIMS, an instrument PC, a notebook or a pipeline, and get back the live and dead counts, aggregates and viability at three sensitivity presets, plus every detection. The counts cover the whole image, so they are not a concentration: cells per mL comes from the app, where the counting region and the dilution are applied. The API documentation is at https://cell-counter-api.fly.dev/v1/docs.
No. CellCount counts from an image of a reusable hemocytometer under the microscope your lab already owns, or from files you already have. There is no benchtop unit, no cassette and no slide to reorder.
How fast is it?
Results come back in seconds for a typical frame.
Can I try it without an account?
Yes. The live demo puts a real sample on the counting grid in your browser and walks you through it square by square. No account, no camera, about a minute.
Most labs start with a call and a demo on their own samples. From there we set up the capture path, a shared counting protocol and an export routine, on site or remote. If you would rather onboard yourself, sign up and follow the public documentation.
The Standard Terms of Service and the Privacy Notice are published on this site, and enterprise arrangements add an individually agreed addendum and order to the standard terms.