Radiomic analysis of postmortem lung changes: a PMCT-based approach for estimating the postmortem interval

Area of Science:

• Forensic Radiology
• Medical Imaging
• Computational Pathology

Background:

• Estimating the postmortem interval (PMI) is critical in forensic medicine.
• Traditional methods for PMI estimation have limitations.
• Quantitative imaging techniques offer potential for improved accuracy.

Purpose of the Study:

• To investigate the utility of radiomic features from postmortem computed tomography (PMCT) scans of the lungs for PMI estimation.
• To identify specific radiomic features that correlate with postmortem changes in the lungs.
• To evaluate the potential of radiomics as a complementary tool in forensic investigations.

Main Methods:

• Sequential PMCT scans were acquired from 17 bodies with known postmortem intervals (4-108 hours).
• Radiomic features were extracted from lung tissues.
• A mixed-effects model was used to analyze the relationship between radiomic features and PMI, testing four model variants.

Main Results:

• Twelve distinct radiomic features showed statistically significant trends with the postmortem interval.
• Features such as cluster shade (GLCM) decreased, median intensity increased, and root mean squared features decreased over time.
• The identified features included first-order statistics, shape, and second-order texture attributes, reflecting lung alterations like gas formation.

Conclusions:

• PMCT-based radiomics shows promise as a complementary tool for postmortem interval estimation.
• Quantitative imaging analysis can enhance the accuracy of forensic investigations.
• Radiomic features provide objective data reflecting postmortem changes in the lungs.