Lipid droplets (LDs) are key markers of cellular metabolism, often altered in cancer. While holotomography enables 3D, label-free imaging of LDs via refractive index, it relies on complex thresholding and lacks biochemical specificity. Here, polarization-sensitive holotomography (PS-HT), which leverages the intrinsic birefringence of LDs for high-contrast, selective identification with a fixed near-zero threshold is presented. Using prostate cell models (healthy PNT2 and cancer PC3), PS-HT is validated against fluorescence microscopy and holotomography, showing that it enables accurate quantification of birefringence, LD volume, dry mass, molecular organization, and spatial distribution. Cancer cells show significantly higher birefringence after glucose treatment, reflecting enhanced lipid accumulation. PS-HT, combined with principal component analysis, achieves near-perfect classification of cancer versus healthy cells, establishing it as a robust, label-free tool for studying lipid metabolism and cancer diagnostics.
Keywords: birefringence imaging; cancer cell metabolism; lipid droplets; polarization‐sensitive holotomography; quantitative phase imaging.
© 2025 The Author(s). Advanced Science published by Wiley‐VCH GmbH.