Algorithms for automatic & non-contact assessment of the cardiovascular system in zebrafish larvae

This paper presents several non-contact methods of assessing the state of the cardiovascular system in zebrafish larvae, based on the videos of the animal under a microscope. A new algorithm for automatic detection of the region of interest (ROI) has been developed. Two distinct approaches are used to convert the information in the video to time series. The first involves using the variation in the intensity, and the second involves multichannel time-series based on color information, both within the ROI. The time series are subsequently analyzed for estimating the heart-rate, and also the components of heart rate variability (HRV). The heart-rate estimated from the algorithms is validated by a comparison with the ground-truth obtained by visual-counting. Power spectrum analysis of the beat-to-beat interval sequence exhibited 2 distinct peaks, in the range of [0.15-0.28] and [0.40-0.6], respectively.