Deep learning predicts zebrafish body size based on local cellular features. ｜Institute of Cellular and Organismic Biology, Academia Sinica

Animal growth is driven by the collective actions of cells, which are reciprocally influenced in real-time by the animal’s overall growth state. While cell behavior and animal growth state are expected to be tightly coupled, it is not yet determined whether local cellular features at the micrometer scale might correlate with the body size of an animal at the macroscopic level. In this study, we demonstrate that analyzing as few as 27 skin cells within a single image of 0.01 mm2 is sufficient to predict the individual’s overall size, ranging from 0.9 to 3.1 mm2. Using a gradient-weighted class activation map (Grad-CAM), we further identified the cellular features influencing the model’s decisions. These results provide a proof-of-concept that deep learning can extract organism-level information from a snapshot of just a few dozen cells. This unexpected finding was published in May this year in Life Science Alliance. The first author, Shang-Ru Yang, is a master’s student in Dr. An-Chi Wei’s laboratory at the Department of Electrical Engineering, National Taiwan University. Co-first author Megan Liaw, is a research assistant in Dr. Chen-Hui Chen’s laboratory at the Institute of Cellular and Organismic Biology (ICOB), Academia Sinica. The study was supported by funding from ICOB, Academia Sinica, National Taiwan University, and the National Science and Technology Council. 

Article title: Deep learning models link local cellular features with whole-animal growth dynamics in zebrafish