Poster
A Graphical User Interface for ML-Based Modeling of Single-Cell and Well-Level Imaging Data
February 5, 2024
•
5 minutes
Spring has developed an ML-based platform targeted towards biologists with no background in machine learning to empower them to analyze cellular imaging data with artificial intelligence. Specifically, we built three modules that are integrated into one cloud-based software application, which we term Supervised Learner, PhenoFinder, and PhenoSorter.
Here, we demonstrate the utility of these three modules by identifying and quantifying different types of cell death. Overall, models trained with only morphological stains perform similarly to models trained with ground truth cell death stains. This opens up the potential of running screening assays and verifying cell death at the same time without compromising use of available imaging channels.
In certain cases, it is possible that dead cells go through some subtle morphological changes before picking up death staining signals. Therefore, using morphological stains to predict cell death can sometimes be more sensitive than using cell death stains.
Our ML tools can be used by scientists without any knowledge in ML to train and apply models on their cellular imaging experiments, providing them with powerful quantitative and predictive analyses.
Authors: Daniel Chen, Kenneth Zhang, Michael Wiest, Ben Komalo, Dat Nguyen, Will Van Trump, Tempest Plott, Elisa Cambronero, Christian Elabd, Rachel DeVay Jacobson
Presenter: Zach Barry