Work

Digital twin reconstruction of red blood cell aging using Raman chemometrics and systems modeling

Year: 2026
Type: preprint
Abstract: Red blood cell (RBC) storage aging involves progressive metabolic and oxidative alterations that compromise membrane integrity and transfusion quality. However, conventional endpoint measurements provide limited information on individual aging dynamics and their predictability. This study presents an experimentally constrained digital twin framework integrating Raman spectral analysis, chemometric modeling, and mechanistic simulation to reconstruct RBC aging trajectories and predict donor-specific hemolysis. Longitudinal Raman spectra and biochemical measurements from packed RBC units stored for eight weeks were analyzed using principal component analysis, least squares decomposition, multivariate curve resolution–alternating least squares, and partial least squares regression. A MATLAB/SimBiology model coupled glucose metabolism, redox regulation, oxidative membrane deterioration, and hemoglobin release. The integrated approach resolved continuous biochemical trajectories characterized by glucose depletion, lactate accumulation, and progressive hemolysis. Despite broadly comparable metabolic kinetics, female-derived RBCs showed slower hemolysis than male-derived RBCs (0.05 ± 0.03 versus 0.40 ± 0.14 week-1). Early Raman spectral features predicted donor-specific hemolysis kinetics (R2 = 0.69), supporting their potential as predictive biomarkers of storage stability. The digital twin reproduced observed metabolic and hemolysis trajectories and implicated hydrogen peroxide regulation, glutathione availability, and GSH/GSSG turnover as candidate contributors to sex-dependent differences. By linking high-dimensional spectroscopic signals with interpretable mechanistic modeling, this framework provides a strategy for tracking RBC storage deterioration and estimating individual aging behavior from early measurements. It may support future development of personalized blood-quality assessment tools.
Source: ChemRxiv
Language: English
Subfield: Hematology
Field: Medicine
Open Access status: green
OpenAlex ID: W7235857519
Cites: 54
Cited by:
Related to: 10