Related Experiment Video
Updated: Aug 12, 2026

LipidUNet-Machine Learning-Based Method of Characterization and Quantification of Lipid Deposits Using iPSC-Derived Retinal Pigment Epithelium
Published on: July 28, 2023
Fluorescence Lifetime Imaging Microscopy Detects Preeclampsia, Gestational Diabetes Mellitus, and Their Comorbidity
Xiaoyu Zhang1,2, Zhongqun Zhan1, Xiaoyu Li2
1Department of Cardiovascular Medicine, Shenzhen University of Advanced Technology General Hospital, Shenzhen518106, China.
Abstract:
Preeclampsia (PE) and gestational diabetes mellitus (GDM) are common but complex pregnancy complications that often co-occur and lack reliable early diagnostic tools. This study introduces a fluorescence lifetime imaging microscopy (FLIM) based approach for detecting metabolic alterations in maternal serum, with Raman spectroscopy and lipidomic analysis providing complementary biochemical context for interpreting group-specific differences. Raman spectroscopy of maternal serum from control, PE, GDM, and PE + GDM groups revealed distinct biochemical alterations, especially in lipid-associated spectral features. Lipidomic validation further identified characteristic metabolic signatures: phosphatidylcholine species were enriched in GDM, while triglycerides were elevated in PE, with PE + GDM displaying overlapping lipid alterations. FLIM captured functional microenvironmental alterations in lipid-associated domains and clearly separated all pathological groups at 24th-28th weeks of gestation, including the comorbid PE + GDM phenotype, a clinically challenging condition for which reliable early indicators are limited. Notably, at 8th-12th weeks of gestation, FLIM detected early biochemical differences through optical signatures, enabling differentiation among disease states at a time when conventional biochemical tests and Raman spectra showed no discernible group differences, achieving strong discrimination performance with area under the curve (AUC) values over 0.8. These results demonstrate the potential of FLIM as a noninvasive, sensitive, and interpretable tool for early risk stratification and longitudinal monitoring of pregnancy complications.