Zebrafish larvae heartbeat detection from body deformation in low resolution and low frequency video

Unlocking the Potential of Zebrafish Research with Artificial Intelligence: Advancements in Tracking, Processing, and Visualization

Zebrafish have become a widely accepted model organism for biomedical research due to their strong cortisol stress response, behavioral strain differences, and sensitivity to both drug treatments and predators. However, experimental zebrafish studies generate substantial data that must be analyzed through objective, accurate, and repeatable analysis methods. Recently, advancements in artificial intelligence (AI) have enabled automated tracking, image recognition, and data analysis, leading to more efficient and insightful investigations. In this review, we examine key AI applications in zebrafish research, including behavior analysis, genomics, and neuroscience. With the development of deep learning technology, AI algorithms have been used to precisely analyze and identify images of zebrafish, enabling automated testing and analysis. By applying AI algorithms in genomics research, researchers have elucidated the relationship between genes and biology, providing a better basis for the development of disease treatments and gene therapies. Additionally, the development of more effective neuroscience tools could help researchers better understand the complex neural networks in the zebrafish brain. In the future, further advancements in AI technology are expected to enable more extensive and in-depth medical research applications in zebrafish, improving our understanding of this important animal model. This review highlights the potential of AI technology in achieving the full potential of zebrafish research by enabling researchers to efficiently track, process, and visualize the outcomes of their experiments.

Using DeepLabCut as a Real-Time and Markerless Tool for Cardiac Physiology Assessment in Zebrafish

Simple Summary

With the advancement of existing technology, artificial intelligence is widely applied in various fields of research, including cardiovascular studies. In this study, we explored the feasibility of conducting a markerless cardiac physiology assessment in zebrafish embryos by using DeepLabCut (DLC), a deep learning tool for motion analysis. Several cardiac parameters, such as heart rate, diastolic–systolic volumes (EDV/ESV), stroke volume, cardiac output, shortening fraction, and ejection fraction were obtained by the DLC-trained model and then compared to the previous published methods, Time Series Analysis and Kymograph. This new method has several advantages, having full automation, precise detection, and real-time labelling. This network was also trained to analyze zebrafish with cardiovascular defects (pericardial edema) induced by chemical treatments with ethanol and ponatinib. It was revealed that the heart rate, EDV/ESV, stroke volume, and cardiac output from both the ethanol and ponatinib groups displayed significant reductions compared with the control. Hopefully, this trained DLC network can contribute to a better understanding and investigation of the existing cardiovascular system and abnormalities.

Abstract

DeepLabCut (DLC) is a deep learning-based tool initially invented for markerless pose estimation in mammals. In this study, we explored the possibility of adopting this tool for conducting markerless cardiac physiology assessment in an important aquatic toxicology model of zebrafish (Danio rerio). Initially, high-definition videography was applied to capture heartbeat information at a frame rate of 30 frames per second (fps). Next, 20 videos from different individuals were used to perform convolutional neural network training by labeling the heart chamber (ventricle) with eight landmarks. Using Residual Network (ResNet) 152, a neural network with 152 convolutional neural network layers with 500,000 iterations, we successfully obtained a trained model that can track the heart chamber in a real-time manner. Later, we validated DLC performance with the previously published ImageJ Time Series Analysis (TSA) and Kymograph (KYM) methods. We also evaluated DLC performance by challenging experimental animals with ethanol and ponatinib to induce cardiac abnormality and heartbeat irregularity. The results showed that DLC is more accurate than the TSA method in several parameters tested. The DLC-trained model also detected the ventricle of zebrafish embryos even in the occurrence of heart abnormalities, such as pericardial edema. We believe that this tool is beneficial for research studies, especially for cardiac physiology assessment in zebrafish embryos.

Quantitative measurements of zebrafish heartrate and heart rate variability: A survey between 1990-2020

Zebrafish is an essential model organism for studying cardiovascular diseases, given its advantages of fast proliferation and high gene homology with humans. Zebrafish embryos/larvae are valuable experimental models used in toxicology studies to analyze drug toxicity, including hepatoxicity, nephrotoxicity and cardiotoxicity, as well as for drug discovery and drug safety screening in the preclinical stage. Heart rate (HR) serves as a functional endpoint in studies of cardiotoxicity, while heart rate variability (HRV) serves as an indicator of cardiac arrhythmia. Cardiotoxicity is a major cause of early and late termination of drug trials, so a more comprehensive understanding of zebrafish HR and HRV is important. This review summarized HR and HRV in a specific range of applications and fields, focusing on zebrafish heartbeat detection procedures, signal analysis technology and well-established commercial software, such as LabVIEW, Rvlpulse, and ZebraLab. We also compared HR detection algorithms and electrocardiography (ECG)-based methods of heart signal extraction. The relationship between HR and HRV was also systematically analyzed; HR was shown to have an inverse correlation with HRV. Applications to drug testing are also highlighted in this review. Furthermore, HR and HRV were shown to be regulated by the automatic nervous system; their connections with ECG measurements are also summarized herein.

An Overview of Methods for Cardiac Rhythm Detection in Zebrafish

The heart is the most important muscular organ of the cardiovascular system, which pumps blood and circulates, supplying oxygen and nutrients to peripheral tissues. Zebrafish have been widely explored in cardiotoxicity research. For example, the zebrafish embryo has been used as a human heart model due to its body transparency, surviving several days without circulation, and facilitating mutant identification to recapitulate human diseases. On the other hand, adult zebrafish can exhibit the amazing regenerative heart muscle capacity, while adult mammalian hearts lack this potential. This review paper offers a brief description of the major methodologies used to detect zebrafish cardiac rhythm at both embryonic and adult stages. The dynamic pixel change method was mostly performed for the embryonic stage. Other techniques, such as kymography, laser confocal microscopy, artificial intelligence, and electrocardiography (ECG) have also been applied to study heartbeat in zebrafish embryos. Nevertheless, ECG is widely used for heartbeat detection in adult zebrafish since ECG waveforms' similarity between zebrafish and humans is prominent. High-frequency ultrasound imaging (echocardiography) and modern electronic sensor tag also have been proposed. Despite the fact that each method has its benefits and limitations, it is proved that zebrafish have become a promising animal model for human cardiovascular disease, drug pharmaceutical, and toxicological research. Using those tools, we conclude that zebrafish behaviors as an excellent small animal model to perform real-time monitoring for the developmental heart process with transparent body appearance, to conduct the in vivo cardiovascular performance and gene function assays, as well as to perform high-throughput/high content drug screening.

High-Throughput Video Processing of Heart Rate Responses in Multiple Wild-type Embryonic Zebrafish per Imaging Field

Heart rate assays in wild-type zebrafish embryos have been limited to analysis of one embryo per video/imaging field. Here we present for the first time a platform for high-throughput derivation of heart rate from multiple zebrafish (Danio rerio) embryos per imaging field, which is capable of quickly processing thousands of videos and ideal for multi-well platforms with multiple fish/well. This approach relies on use of 2-day post fertilization wild-type embryos, and uses only bright-field imaging, circumventing requirement for anesthesia or restraint, costly software/hardware, or fluorescently-labeled animals. Our original scripts (1) locate the heart and record pixel intensity fluctuations generated by each cardiac cycle using a robust image processing routine, and (2) process intensity data to derive heart rate. To demonstrate assay utility, we exposed embryos to the drugs epinephrine and clonidine, which increased or decreased heart rate, respectively. Exposure to organic extracts of air pollution-derived particulate matter, including diesel or biodiesel exhausts, or wood smoke, all complex environmental mixtures, decreased heart rate to varying degrees. Comparison against an established lower-throughput method indicated robust assay fidelity. As all code and executable files are publicly available, this approach may expedite cardiotoxicity screening of compounds as diverse as small molecule drugs and complex chemical mixtures.

Anisotropic organization of circumferential actomyosin characterizes hematopoietic stem cells emergence in the zebrafish

Hematopoiesis leads to the formation of blood and immune cells. Hematopoietic stem cells emerge during development, from vascular components, via a process called the endothelial-to-hematopoietic transition (EHT). Here, we reveal essential biomechanical features of the EHT, using the zebrafish embryo imaged at unprecedented spatio-temporal resolution and an algorithm to unwrap the aorta into 2D-cartography. We show that the transition involves anisotropic contraction along the antero-posterior axis, with heterogenous organization of contractile circumferential actomyosin. The biomechanics of the contraction is oscillatory, with unusually long periods in comparison to other apical constriction mechanisms described so far in morphogenesis, and is supported by the anisotropic reinforcement of junctional contacts. Finally, we show that abrogation of blood flow impairs the actin cytoskeleton, the morphodynamics of EHT cells, and the orientation of the emergence. Overall, our results underline the peculiarities of the EHT biomechanics and the influence of the mechanical forces exerted by blood flow.

As humans, we have two major types of blood cell: our red blood cells transport oxygen around the body, while our white blood cells fight disease. Both types of cell come from the same stem cells, which first appear early in embryonic development. These stem cells emerge from the walls of major blood vessels, including the aorta – which carries blood from the heart.

Stem cells have not yet decided which adult cell to become. Given the right signals, blood stem cells can form red blood cells or any of the different types of white blood cell. Understanding this process could allow scientists to recreate it in the laboratory, making blood stem cells that can give rise to all blood cells found in the body. But, this is not yet possible because we do not know all the conditions needed to make the cells and ensure they survive. One crucial gap in our understanding concerns the importance of blood flow. As the main blood vessel leaving the heart, the aorta experiences mechanical stress every time the heart beats. Lancino et al. wanted to find out whether this influences the development of the blood stem cells.

Zebrafish embryos are transparent, making it easy to see their bodies developing under a microscope. Like humans, they also produce both red blood cells and white blood cells meaning Lancino et al. could watch the birth of blood stem cells in these embryos from a part of the aorta called the aortic floor. A new software tool unwrapped pictures of the tube-shaped blood vessel into flat, two-dimensional maps, making it possible to see how the aorta changed over time. This revealed that, as blood stem cells leave the aortic floor, they bend and contract with the direction of the blood flow. Rings of actin and myosin proteins that formed around the stem cells as they are born helped the process along, while stopping the heartbeat changed the way the blood cells emerged. Without any blood flow, the actin proteins did not assemble properly; the stem cells also emerged in the wrong direction and some of them even died.

These findings show that physical forces play a role in the formation of blood stem cells. Understanding this process brings scientists a step closer to recreating the conditions for making different kinds of blood cells outside of the body.