Bringing Artificial Intelligence (AI) into Environmental Toxicology Studies: A Perspective of AI-Enabled Zebrafish High-Throughput Screening.

The booming development of artificial intelligence (AI) has brought excitement to many research fields that could benefit from its big data analysis capability for causative relationship establishment and knowledge generation. In toxicology studies using zebrafish, the microscopic images and videos that illustrate the developmental stages, phenotypic morphologies, and animal behaviors possess great potential to facilitate rapid hazard assessment and dissection of the toxicity mechanism of environmental pollutants. However, the traditional manual observation approach is both labor-intensive and time-consuming. In this Perspective, we aim to summarize the current AI-enabled image and video analysis tools to realize the full potential of AI. For image analysis, AI-based tools allow fast and objective determination of morphological features and extraction of quantitative information from images of various sorts. The advantages of providing accurate and reproducible results while avoiding human intervention play a critical role in speeding up the screening process. For video analysis, AI-based tools enable the tracking of dynamic changes in both microscopic cellular events and macroscopic animal behaviors. The subtle changes revealed by video analysis could serve as sensitive indicators of adverse outcomes. With AI-based toxicity analysis in its infancy, exciting developments and applications are expected to appear in the years to come.

Deep Learning-Enabled Morphometric Analysis for Toxicity Screening Using Zebrafish Larvae.

Toxicology studies heavily rely on morphometric analysis to detect abnormalities and diagnose disease processes. The emergence of ever-increasing varieties of environmental pollutants makes it difficult to perform timely assessments, especially using in vivo models. Herein, we propose a deep learning-based morphometric analysis (DLMA) to quantitatively identify eight abnormal phenotypes (head hemorrhage, jaw malformation, uninflated swim bladder, pericardial edema, yolk edema, bent spine, dead, unhatched) and eight vital organ features (eye, head, jaw, heart, yolk, swim bladder, body length, and curvature) of zebrafish larvae. A data set composed of 2532 bright-field micrographs of zebrafish larvae at 120 h post fertilization was generated from toxicity screening of three categories of chemicals, i.e., endocrine disruptors (perfluorooctanesulfonate and bisphenol A), heavy metals (CdCl2 and PbI2), and emerging organic pollutants (acetaminophen, 2,7-dibromocarbazole, 3-monobromocarbazo, 3,6-dibromocarbazole, and 1,3,6,8-tetrabromocarbazo). Two typical deep learning models, one-stage and two-stage models (TensorMask, Mask R-CNN), were trained to implement phenotypic feature classification and segmentation. The accuracy was statistically validated with a mean average precision >0.93 in unlabeled data sets and a mean accuracy >0.86 in previously published data sets. Such a method effectively enables subjective morphometric analysis of zebrafish larvae to achieve efficient hazard identification of both chemicals and environmental pollutants.