CountAnywhere: frugal cell counting and viability from smartphone hemocytometer photos
Cell counting sets the concentration, viability and seeding density of most cell-based experiments, yet it still relies on a slow, operator-dependent hemocytometer count or on an automated counter that needs a dedicated instrument and a disposable slide for every sample. Here we present CountAnywhere, a browser-based platform that counts cells and scores trypan-blue viability from an ordinary…
CountAnywhere is a novel platform that offers rapid and accurate cell counting and viability assessment from smartphone images of hemocytometer slides. Traditionally, cell counting and viability determination required laborious hemocytometer counting or expensive automated counters, but CountAnywhere simplifies the process with just a standard microscope, hemocytometer, and a smartphone. The platform is accessible through a web app or a local server and requires no additional hardware.
CountAnywhere relies on the image's built-in counting grid as a ruler, focus check, and stereological counting frame. It employs the Cellpose-SAM model, which has been fine-tuned on five cell lines, resulting in a significant improvement in segmentation accuracy. The F1 score increased from 0.46 to 0.97, demonstrating the model's exceptional performance in capturing both live and dead cells with remarkable precision.
To classify cell viability, CountAnywhere utilizes a local-contrast rule, which demands no fitted parameters. This approach successfully categorized 5,739 hand-labeled cells as either alive or dead, achieving an impressive 98% sensitivity and 99% specificity. When compared to manual cell counts, the app's counts exhibited an even closer alignment with the ground truth (concordance of 0.999) than manual counts (0.970), along with a more consistent performance as cell density increased (mean absolute error of 5.3% versus 10.7% for manual counts).
The efficiency of CountAnywhere is another notable advantage. A full chamber scan takes approximately 50 seconds on the app, whereas manual counting requires about 122 seconds. This speed and accuracy make CountAnywhere an invaluable tool for researchers, democratizing cell counting and making it accessible to a broader range of users.
Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
