From biochemical markers to machine learning: a review of postmortem interval estimation via vitreous humor analysis

Accurate estimation of the postmortem interval (PMI) is pivotal in forensic investigations and medicolegal proceedings. However, conventional estimation methods are often compromised by environmental factors and variable preservation conditions, restricting their accuracy and practical utility. To overcome these challenges, vitreous humor (VH), sequestered by the blood-retinal barrier (BRB), exhibits relative metabolic stability and resistance to decomposition, rendering it an optimal biological matrix for PMI estimation. In this review, we critically evaluate the diagnostic utility of VH biochemical constituents for PMI estimation, comprehensively examining the correlation between post-mortem compositional alterations in VH and the PMI. By synthesizing findings from prior studies, we demonstrate how VH significantly expands the effective temporal window for precise PMI estimation. Furthermore, we elucidate the application of machine learning models in characterizing the complex dependencies between VH constituents and the PMI, which not only enhances predictive accuracy but also facilitates robust nonlinear time series analysis. Finally, we outline future perspectives, envisioning a multimodal fusion framework for VH analysis that is poised to emerge as an indispensable tool for PMI estimation.