Morphological characteristics of zebrafish's yolk sac for malformation based on orthogonal‐polarization‐gating optical coherence tomography

In this study, an automatic algorithm combining an ellipsoid approximation and U‐net has been presented for the characterization of a zebrafish's yolk sac. The polarization‐difference‐balanced‐detection image of zebrafish was obtained based on orthogonal‐polarization‐gating optical coherence tomography and used to segment the yolk sac region. And ellipsoid can approximate the shape of the three‐dimensional yolk sac, and the multiple parameters of volume and the three principal axes (k, l and m) can be used to quantify the yolk sac. In addition, the multiple parameters of two principal axes (l and m) and volume can distinguish the malformation from the normal controlled group. Finally, the volume malformation of the yolk sac calculated by the proposed algorithm ranges from 16.55% to 46.05%. Thus, the degree of malformation can be applied for toxicity analysis. And this method provides a potential application for an accurate judgment index for biotoxicological testing. The polarization‐difference‐balanced‐detection image of zebrafish was obtained based on orthogonal‐polarization‐gating optical coherence tomography (OPG‐OCT). An automatic algorithm combining an ellipsoid and a convolutional neural network (CNN) employing U‐net has been presented for approximation of a zebrafish's yolk sac.