Runge, T (2023) The Use of Maxillary Sinus Imaging as a Tool in Human Identification. Doctoral thesis, Liverpool John Moores University.

Preview

Text 2023RungePhD.pdf - Published Version Available under License Creative Commons Attribution Non-commercial. Download (8MB) | Preview

Abstract

The presented research performs evaluations of maxillary sinus morphologies for human identification purpose. Due to the durability of maxillary sinuses, morphological analyses of the structure can prove to be extremely valuable and informative, even when parts of the skull are destroyed and dental records cannot be applied. This research is comprised of 4 studies evaluating maxillary sinus morphologies in order to build a comprehensive outlook on the methodological potentials. In total the used sample is comprised of right and left maxillary sinuses from 988 individuals divided into 12 populations. The morphologies are assessed by extracting the maxillary sinuses from radiographic and CT images and applying elliptic Fourier analyses on the structures. Morphological variability is investigated by converting the maxillary sinus morphology into multiple closed curves, and embedding them into a cartesian system. Principal component analyses on four components further simplifies the processing.

The first two studies of this research are concerned with morphological uniqueness testing both in a simulated and real-life scenario to lay a comprehensive foundation for method applicability. Uniqueness testing is executed as a morphological ante- and postmortem comparison by calculating Euclidean and Mahalanobis distances. Euclidean correlation values from 0.000 in the simulated sample up to 0.002 in the real-life sample indicate maxillary sinus morphological uniqueness for each ante- and postmortem sinus morphology pair. Mahalanobis distances are used for visualisation. The third study is assessing the reproducibility of maxillary sinus morphological extraction by applying Cohen’s kappa values. The high kappa values in intra- and inter-observer reliability testing indicate high quality extraction and interpretation of morphologies, increasing the methodological confidence level. Finally, the last study is dedicated to understanding age-related changes in maxillary sinuses by calculating growth rates by population and by sex on Euclidean distances. All evaluations reveal quasi-linear and monotonously rising distances with growth rates varying among left and right sinuses and population.

This research advances the potential of maxillary sinus morphologies for human identification and demonstrates its advantages over other paranasal identification methodologies. Therefore, this research acts as an essential first step toward using the proposed methodological framework in future forensic casework.

Item Type:	Thesis (Doctoral)
Uncontrolled Keywords:	Maxillary Sinus; Human Identification; Geometric Morphometrics; Radiography
Subjects:	Q Science > QH Natural history > QH301 Biology Q Science > QM Human anatomy Q Science > QP Physiology
Divisions:	Biological & Environmental Sciences (from Sep 19)
SWORD Depositor:	A Symplectic
Date Deposited:	04 Jul 2023 10:15
Last Modified:	04 Jul 2023 10:16
DOI or ID number:	10.24377/LJMU.t.00020015
Supervisors:	Borrini, M, Irish, J and Ventresca Miller, A
URI:	https://researchonline.ljmu.ac.uk/id/eprint/20015

View Item