biology cell imaging fluorescence imaging high-throughput imaging imaging life sciences machine learning microscopy
Quantifying cell morphology using images and machine learning models has proven to be a powerful tool to study the response of cells to treatments. However, the models used to quantify cellular morphology are typically trained with a single microscopy imaging type and under controlled experimental conditions. This results in specialized models that cannot be reused across biological studies because the technical specifications do not match (e.g., different number of channels), or because the target experimental conditions are out of distribution. We have created CHAMMI-75, a large-scale dataset containing 2.8 million multi-channel, high-resolution images curated from 75 diverse, publicly available biological studies. This dataset is useful to investigate and develop channel-adaptive models, which could process microscopy images of varying technical specifications and regardless of the number of channels. By breaking the limitations of existing models, CHAMMI-75 is an invaluable resource for creating the next generation of foundation models for image-based biological research.
Every 2 years
CC BY 4.0 License
https://github.com/CaicedoLab/CHAMMI-75
Morgridge Institute for Research
See all datasets managed by Morgridge Institute for Research.
Juan Caicedo, juan.caicedo@wisc.edu
CHAMMI-75 was accessed on DATE from https://registry.opendata.aws/chammi.
arn:aws:s3:::chammi-dataus-west-2aws s3 ls --no-sign-request s3://chammi-data/