Fully automatic CT-histogram-based fat estimation in dead bodies.

Area of Science:

• Forensic Science
• Medical Imaging
• Computational Anatomy

Background:

• Temperature-based death time estimations (TDE) rely on post-mortem body cooling models.
• Current models require accurate quantification of individual parameters like abdominal fat.
• Existing methods lack precision in assessing abdominal fat distribution, hindering TDE accuracy.

Purpose of the Study:

• To develop and validate an automated algorithm for precise quantification and spatial distribution analysis of abdominal fat using computed tomography (CT) data.
• To improve the accuracy of physical post-mortem body cooling models for time of death estimations.

Main Methods:

• Development of a CT-histogram-based fat estimation and quasi-segmentation (CFES) algorithm.
• CFES utilizes a weighted least squares method on CT-slice histograms, requiring no anatomical prior information or feature detection.
• Validation performed on synthetic data, a specialized phantom, and a deceased individual's CT scan.

Main Results:

• CFES accurately quantifies abdominal body fat, muscle, organ, and connective tissue from CT slices.
• The algorithm generates a quasi-segmentation differentiating fat from other tissues.
• Anatomically plausible results were obtained when applied to a post-mortem CT scan.

Conclusions:

• CFES offers a precise, automated method for abdominal fat quantification using CT imaging.
• This algorithm can significantly enhance the accuracy of physical models used in forensic time of death estimations.
• Automated CT analysis provides a valuable tool for forensic science in determining time of death.