Anesthesia-free Heartbeat Measurements in Freely Moving Zebrafish

This study aims to monitor zebrafish heartbeat completely un-invasively without anesthesia, visible light, or other stress factors that may influence cardiac function. Fast and almost unpredictable,, fish motion is the main experimental change here. A significant achievement of our lab is the development of a neural network algorithm capable of precisely tracking of zebrafish heart at up to 100 frames per second. This project addresses two significant gaps, the short period of optical zebrafish studies, which limited to five to seven days post fertilization and the necessity of anesthesia. This protocol utilizes shortwave infrared illumination, making zebrafish transparent up to two months post fertilization, along with an all-image processing technique to track the zebrafish heart.

[Instructor] To begin, connect the camera to the computer, launch the camera application, and enable the preview mode. Turn on the illuminator. Using a Pasteur pipette, place the zebrafish larvae in an agarose mold. Position the agarose mold on the stage. Adjust the exposure time and illuminator power to ensure almost complete illumination of the larva's head, keeping the exposure time below 1.5 milliseconds. For image acquisition, set the image bit depth to 12 bits and the frame rate to at least 60 frames per second. Then set the number of frames so that images are acquired for at least 10 seconds. Set the frame naming format and file type. Start acquiring images. To prepare dataset for neural network training, launch Matlab. Open the script file DatasetPreparation.m. Click run to run the script. In the popup window, select the directory containing the outlined images, the folder PixelLabelData, and the file gtruth.mat. After acquiring the images, launch the AutoHR application. Navigate to the processing tab and click choose models to import trained neural networks. Click choose directory to import acquired images for analysis, click process to begin the analysis. The heart rate of zebrafish larvae at 12 days post fertilization was determined using photoplethysmography or PPG. High quality frames with a heart area diameter of 20 pixels resulted in a clear PPG signal and Fourier analysis identified a peak frequency corresponding to a heart rate of 130 beats per minute. Low quality frames with a heart area diameter of 14 pixels introduced significant noise in the PPG signal, reducing measurement reliability. The method also detected an increased heart rate of 170 beats per minute. The validation protocol demonstrated that heart rate significantly increased following the application of a provocation stimulus. The median heart rate rose from approximately 122 beats per minute to 155 beats per minute.