Multimodal Approaches Based on Microbial Data for Accurate Postmortem Interval Estimation.

Area of Science:

• Forensic Science
• Microbiology
• Biochemistry

Background:

• Accurate postmortem interval (PMI) estimation is vital for forensic investigations and legal proceedings.
• Early PMI indicators (body temperature, rigor mortis) become unreliable as decomposition advances.
• Decomposition involves predictable microbial activity, offering potential for objective PMI assessment.

Purpose of the Study:

• To explore microbial community shifts during decomposition as objective criteria for PMI estimation.
• To investigate the transition from anaerobic to aerobic microbiomes and its relation to decomposition stages.
• To evaluate the utility of multi-modal data, including microbial, metabolic, and volatile organic compound (VOC) profiles, for late-stage PMI determination.

Main Methods:

• Analyzing microbial community succession in different organs and skin/soil environments.
• Characterizing metabolic and volatile organic compound (VOC) profiles during decomposition.
• Applying machine learning (ML) models to integrate multimodal data for PMI prediction.

Main Results:

• Significant shifts in microbial communities (anaerobic to aerobic) correlate with decomposition stages (pre- and post-rupture).
• Distinct microbial successions observed in different organs provide specific PMI insights.
• Multi-modal data integration using ML shows promise for enhancing PMI estimation accuracy.

Conclusions:

• Microbial analysis and multi-modal data integration represent a significant advancement for late-stage PMI estimation.
• Machine learning models are crucial for effectively utilizing complex, multimodal datasets in forensic science.
• Future research should focus on human-specific databases and microbial-insect interactions to further refine PMI accuracy.