OpenBloodFlow: A User-Friendly OpenCV-Based Software Package for Blood Flow Velocity and Blood Cell Count Measurement for Fish Embryos

Simple Summary Artificial intelligence (AI) has been established to contribute in number of research areas such as medical imaging, diagnostic tools, ultrasound, cardiac scans, X-rays and blood flow analysis. OpenCV is a library having programming functions that provide ample of modules for digital image processing and artificial intelligence. In this study, we introduced a software package as OpenBloodFlow that can measure blood flow velocity and blood cell count precisely by selecting the dorsal aorta of zebrafish. The program is based on python programing language, which is a high-level, general-purpose language to solve biological problems. We present a cost-effective and fully automatic tool to perform the analysis as compared to commercially available software(s) that are not freely available to access. The key features of the program include a user-friendly graphical user interface (GUI), blood flow velocity and blood cell count in the given video dataset. The results are automatically saved in a CSV file without using any external plugins or third-party software and the program does not require GPU average CPU can be used for the execution. The transparent appearance of fish embryos provides an excellent assessment feature for observing cardiovascular function in vivo. Previously, methods to conduct vascular function assessment were based on measuring blood-flow velocity using third-party software. In this study, we reported a simple software, free of costs and skills, called OpenBloodFlow, which can measure blood flow velocity and count blood cells in fish embryos for the first time. First, videos captured by high-speed CCD were processed for better image stabilization and contrast. Next, the optical flow of moving objects was extracted from the non-moving background in a frame-by-frame manner. Finally, blood flow velocity was calculated by the Gunner Farneback algorithm in Python. Data validation with zebrafish and medaka embryos in OpenBloodFlow was consistent with our previously published ImageJ-based method. We demonstrated consistent blood flow alterations by either OpenBloodFlow or ImageJ in the dorsal aorta of zebrafish embryos when exposed to either phenylhydrazine or ractopamine. In addition, we validated that OpenBloodFlow was able to conduct precise blood cell counting. In this study, we provide an easy and fully automatic programming for blood flow velocity calculation and blood cell counting that is useful for toxicology and pharmacology studies in fish.