2025 Volume 40 Issue 1 Pages 21-30
Understanding the spatial progression of liver fibrosis is crucial for evaluating the disease severity and therapeutic response. In this study, we developed a method for reconstructing three-dimensional (3D) histological images of fibrotic liver tissue from serial Azan-stained sections in a nonalcoholic steatohepatitis model rat. Quantitative analysis of the reconstructed volumes revealed a time-dependent increase in fibrotic areas. We applied mathematical growth models to describe the progression of fibrosis over time. Among several candidates, the Gompertz model was favored based on the Akaike Information Criterion, its corrected version, and the Bayesian Information Criterion, whereas the Root Mean Square Error suggested that the Generalized Logistic model had the best fit. Compared with random two dimensional sampling, 3D reconstructions showed reduced variability in fibrosis estimates, thereby supporting their robustness. These results demonstrate the utility of 3D reconstruction combined with statistical modeling for capturing fibrotic dynamics and offer a foundation for future diagnostic and therapeutic approaches.
