This invention discloses a four-dimensional Raman
monitoring system and dynamic quantitative method for environmental
microplastics in vivo, belonging to the field of environmental
ecological monitoring and analysis technology. The
system includes a
confocal spectroscopy acquisition module, a controller, and a
data processing module. It acquires the three-dimensional coordinates of live samples through coordinated spatial scanning and simultaneously acquires Raman signals to construct four-dimensional
spectral data. A multidimensional convolutional instance segmentation network is used to process the four-dimensional Raman
spectral data, identifying and classifying endogenous biological matrices and exogenous multi-mixed
microplastics in vivo, outputting
voxel-level semantic classification results, and then quantitatively calculating the absolute retention volume and abundance of
microplastics in the
organism. This invention overcomes the shortcomings of existing technologies in
label-free quantification under complex biological backgrounds, achieving in-situ accurate identification of mixed microplastics
in vivo, and providing a standardized quantitative monitoring method for assessing the
bioaccumulation, metabolic patterns, and ecotoxicological effects of environmental microplastics.