Personal identification using frontal sinus coding methods: The effect of mixed image modality comparisons

Area of Science:

• Forensic Anthropology
• Radiology
• Biometrics

Background:

• Code-based methods for frontal sinus comparison exist for skeletal identification.
• Limited data available on matched-pair accuracy and impact of imaging modalities.

Purpose of the Study:

• Validate exclusion rates of Cameriere et al. and Tatlisumak et al. coding methods.
• Establish matched-pair accuracy rates for these methods.
• Assess consistency of individual variables across different image modalities.

Main Methods:

• Examined frontal sinus coding methods (Cameriere et al., Tatlisumak et al.) in 225 individuals (African American, Native American, European American).
• Compared scoring consistency of individual variables and string codes between radiographs and CT-based 3D models.
• Performed interobserver and intraobserver analyses.

Main Results:

• Exclusion rates were high (93%-96%), consistent with prior studies.
• Matched-pair accuracy rates were low (13%-18%).
• Arcades showed poor scoring consistency (p < 0.001) between modalities; supraorbital cells had minimal inter/intraobserver agreement.

Conclusions:

• Current frontal sinus coding methods are not recommended for independent use in skeletal identification due to low accuracy and variable consistency.
• Visual identification remains crucial for inclusion/exclusion when analyzing frontal sinuses.
• Demographic factors did not influence match probability.