FEHAT: efficient, large scale and automated heartbeat detection in Medaka fish embryos

Area of Science:

• Developmental biology
• Bioinformatics
• Physiological monitoring

Background:

• High-resolution imaging of model organisms enables physiological measurement.
• Transparent fish embryos allow visualization of key processes like heartbeat.
• High-throughput systems generate large datasets for developmental studies.

Purpose of the Study:

• To develop an automated software solution for analyzing large-scale heart rate imaging datasets from Medaka fish embryos.
• To address the need for efficient analytical schemes for high-throughput video data.
• To enable robust investigation of developmental processes and system perturbations.

Main Methods:

• Developed FEHAT software for automated analysis of Medaka fish embryo heart rate videos.
• Utilized image segmentation to identify embryo regions with pixel intensity changes.
• Employed heart classification and Fourier Transformations for heart rate estimation.
• Tested FEHAT on large datasets of embryos in 96-well plates.

Main Results:

• FEHAT provides automated, high-throughput analysis of thousands of heart rate imaging datasets.
• The software reliably estimates heart rates across various temperature perturbations.
• Performance was compared to established, less automated solutions, demonstrating FEHAT's efficiency.
• FEHAT requires no human intervention for analyzing tens of thousands of embryos.

Conclusions:

• FEHAT offers a fully automated solution for analyzing large numbers of fish embryo heart rate videos.
• This tool facilitates robust investigation of developmental biology and physiological responses.
• FEHAT's high-throughput capability supports large-scale studies without manual oversight.