Multiscale and recursive unmixing of spatiotemporal rhythms for live-cell and intravital cardiac microscopy

Cardiovascular diseases represent a significant public health challenge, highlighting the urgent need for effective therapeutic strategies based on a comprehensive understanding of cardiac development. While fluorescence microscopy is an invaluable tool for probing biological processes, cardiovascular signals are often complicated by persistent autofluorescence, overlaying dynamic cardiovascular entities, and non-specific labeling from tissue microenvironments. In response, this study presents multi-scale recursive decomposition (MSR), an algorithm for the precise extraction of dynamic cardiovascular signals. MSR constructs a comprehensive framework for cardiac microscopy that includes pixelwise image enhancement, robust principal component analysis, and recursive motion segmentation. This method has been validated in the in vivo studies of cardiovascular morphology in <i>Xenopus</i> embryos. We anticipate the methodology to advance cardiovascular study across a broad spectrum of cardiac models.