AI Cell Quantification

CellQuant 1.2.4. Counts nuclei (antibody positive or negative), nuclei containing a red punctum ("Nostril") and puncta inside or outside nuclei in RGB fluorescence images (blue = DAPI, green = antibody, red = puncta). Trained on CVAT outlines of 35-day spinal-cord sections stained for Iba1 or TMEM119.
Cell, Antibody +Cell, Antibody -Cell, Nostril, Antibody +Cell, Nostril, Antibody -Nostril in NucleusNostril External of Nucleus

1. Analyse your images

Every image is analysed as 8 rotations and flips and the outputs averaged (the setting that scored best in cross-validation). A written summary follows the counts.
0.1 0.9
0.1 0.9
10 300
Punctum type from: network head / position relative to nuclei
Nucleus gets 'Nostril' if: contains an in-nucleus punctum / that OR the network head / head only
Example (512 px crop of a training image)

Counts

2. Check the model against the 9 annotated images

Click one of the 9 images. The model runs on it (8 rotations and flips, averaged) and three panels appear: the manual annotation, the model prediction (both in the label colours above) and missed only, the annotated objects the model did not find, outlined in red with their row number from the list at the bottom. The final model was trained on these images, so this shows its fit, not its accuracy on new images (those numbers are in the README). A written summary of the comparison follows the panels.

No image selected yet.

Open a panel full screen with the button in its corner. Matched: found with the same label. Wrong label: found, but given another label. Missed: annotated, not found. Extra: found, not annotated. Difference = predicted − annotated. Nuclei match when their outlines overlap with IoU ≥ 0.5, puncta when their centres are within 6 px.

Annotated vs predicted, per label

Every missed, extra or wrongly labelled object, numbered as on the panels (x, y in pixels from the top-left corner)

Nucleus outlines come from a watershed over the network's distance and class maps. A nucleus is labelled Nostril when a punctum classified as in-nucleus lies inside it. Images larger than 1536 px are processed in overlapping tiles. Nothing leaves the Space except, the counts and one overlay thumbnail sent to Gemini for the written summary (only when the Space has a Gemini key). Inference runs on a shared GPU (Hugging Face ZeroGPU): no account is needed, each visitor has a daily allowance of GPU minutes and one image takes a few seconds.