A novel method for determining postmortem interval based on the metabolomics of multiple organs combined with...

Area of Science:

• Forensic Science
• Biochemistry
• Computational Biology

Background:

• Accurate postmortem interval (PMI) determination is vital in forensic investigations.
• Current PMI estimation methods have limitations, and machine learning approaches are emerging.
• Multi-organ analysis for PMI estimation using compound profiles is an underdeveloped area.

Purpose of the Study:

• To develop and evaluate a multi-organ stacking model for estimating PMI.
• To analyze differential compounds from rat skeletal muscle, liver, lung, and kidney.
• To compare the performance of single-organ and multi-organ models.

Main Methods:

• Collected tissue samples (skeletal muscle, liver, lung, kidney) from 140 rats at various time points postmortem.
• Utilized ultra-performance liquid chromatography-high-resolution mass spectrometry for compound profiling.
• Applied multivariate statistical analysis and developed single-organ and multi-organ stacking machine learning models.

Main Results:

• The multi-organ stacking model demonstrated the highest accuracy (93%) and area under the receiver operating characteristic curve (0.96).
• External validation showed high reliability, with only 1 out of 14 samples misclassified.
• Discriminant compounds from multiple organs significantly improved PMI estimation accuracy.

Conclusions:

• A multi-organ stacking model integrating compound data from multiple organs offers a promising advancement in forensic PMI estimation.
• This approach enhances accuracy and reliability compared to single-organ models.
• The developed model represents a potential novel forensic tool for precise PMI determination.