Unravelling geometric morphometric method on frontal sinus for biological profiling among Malaysian from radiograph: a novel approach in forensic anthropology

Choy Ker Woon1,2, Nur Damia Iwani Zulkiflee1, Aspalilah Alias3, Helmi Mohd Hadi Pritam4, Mansharan Kaur Chanchal Singh2, Eric Chung5

1Department of Anatomy, Faculty of Medicine, Universiti Teknologi MARA, Selangor, Malaysia; 2Institute of Pathology, Laboratory and Forensic Medicine (I-PPerForM), Faculty of Medicine, Universiti Teknologi MARA, Selangor, Malaysia; 3Department of Basic Sciences and Oral Biology, Faculty of Dentistry, Universiti Sains Islam Malaysia, Malaysia; 4Department of Forensic Science, Faculty of Health Sciences, Universiti Sains Malaysia, Kelantan, Malaysia; 5Department of Biomedical Imaging, Faculty of Medicine, University of Malaya, Malaysia

Objectives: Biological profiling is necessary for skeletonised remains. The traditionally used linear measurement has low reliability and size-oriented, whereas geometric morphometric (GM) is reliable in characterising the shape. However, studies of frontal sinus (FS) among Malaysians using GM method are unexplored. This study aims to evaluate FS pattern distribution and morphology differences of FS between sex, race and age among Malaysians using GM as a novel approach.
Methods: Skull radiographs of 504 adult Malaysians comprising males and females of Malay, Chinese and Indian races aged of ≥20 years old, were retrieved from University Malaya Medical Centre. FS patterns were descriptively classified according to its existence, symmetrical, dominance and lobulations. Then, eight landmarks were applied to the radiographs using TPSDig2 software. GM analysis including Principal component (PC) analysis, Procrustes analysis of variance ANOVA, canonical variates analysis (CVA) and discriminant function analysis (DFA), were performed using MorphoJ software.
Results: Bilateral presence of FS is common (95.4%). The first three PC exhibited 83.0% variation. ANOVA revealed significant differences in centroid size and shape of FS between different sex. DFA n demonstrated average classification accuracy of 80.6% for sex estimation, 57.4% for race estimation and 53.5% for age estimation.
Conclusion: Pattern distribution of FS is unique between individuals. The absence of FS is rare, making it useful for forensic identification. FS is a valuable tool for sex identification, but not reliable for race and age estimation among Malaysians. This is the first such study looking at the FS for profiling amongst Malaysian population.

Keywords: biological profiling; forensic anthropology; frontal sinus; geometric morphometric; malaysian population

Ethical statement: UMMC-Medical Research Ethics Committee (MREC) (MECID No: 2022119 – 10937).

Funding statement: Faculty of Medicine Research Grant- 600-TNCPI 5/3/DDF (MEDIC) (004/2023)