Most human embryos are mosaic for chromosome abnormalities: containing cells that are euploid and aneuploid. Currently, a cell biopsy from the trophectoderm lineage of the blastocyst embryo is used to diagnose aneuploidy. However, this does not provide a diagnosis of the proportion of aneuploid cells in the remaining trophectoderm cells or within the inner cell mass (cells that form the fetus). Hence, the development of a non-invasive tool to determine the proportion of aneuploid cells would be clinically valuable. Aneuploidy in human embryos leads to altered metabolism. Co-factors utilised in cellular metabolism are autofluorescent and can be used to predict the metabolic state of cells. We hypothesised that aneuploid cells within the preimplantation embryo could be non-invasively discerned by their autofluorescent spectra. Using primary human fibroblast cells with known aneuploidies and a mouse embryo model with differing ratios of euploid:aneuploid cells we investigated whether we could distinguish euploid from aneuploid using hyperspectral imaging. Hyperspectral imaging of 1:1 (euploid:aneuploid) chimeric embryos showed a distinct spectral profile compared to euploid embryos. Following unsupervised linear unmixing, the abundance of FAD in the inner cell mass of aneuploid blastocysts was significantly lower compared to euploid blastocysts. For human fibroblasts, we were able to clearly distinguish between euploid and aneuploid with different karyotypes. We demonstrate that hyperspectral imaging is able to distinguish cells based on their ploidy status making it a promising tool in assessing embryo mosaicism.