ASSESSMENT OF SEXUAL DIMORPHISM FROM THE MASTOID TRIANGLE USING 3D CT SCAN IN NEPALESE POPULATION

1* 2 3 4 5 Pankaj Kumar Singh , Raj Kumar Karki , Abdul Sami Khan , Dhiraj Kumar Shah , Sushant Bhardwaj Introduc on Skull plays an important role, second only next to pelvis for sex determina on. In skull, mastoid and occipital region due to its anatomical loca on is the most protected region, offering high resistance to decomposi on and any form of tempering. With the intension of studying sexual dimorphism in the mastoid region involving mastoid process, we ini ated this study. The technique we adopted was first introduced by Paiva and Segre, i.e. measuring the distance between three cranio-metric land mark (Porion, Mastoidale, and Asterion) and calcula ng the area of the triangle to study sexual dimorphism. Objec ve The objec ve of this study is to determine sex based on mastoid triangle dimensions measured in 3D reconstructed computed tomography of the skull. Methodology The study included 196, 3D reconstructed CT scan images of skull, 98 of which were males and 98 of which were females. "The three craniometric points were marked to demarcate the mastoid triangle on both side". Heron's formula was used for measurement of mastoid triangle areas. Result Descrip ve sta s cs and the student t-test revealed that males have larger mastoid triangle dimensions in all measurements than females, with a significant difference (p<0.000) between them. The paired t-test revealed no sta s cally significant difference (p<0.05) between the right and le sides. Apart from angle asterion (p<0.05) on the right side, no other mastoid triangle angles on either side showed a significant difference. The predictability of mastoid triangle measurements and areas was determined using a discriminant func on and ROC curve analysis. The calculated mastoid triangle measurements and areas were 72.5 % of Asterion to Porion, 74.5 % of Asterion to Mastoidale, 86.3 % of Mastoidale to Porion, and 88.2 % of areas, respec vely. Conclusion The study findings suggest that mastoid triangle could be used to determine gender in the case of fragmentary skull remains by the forensic experts in the course of inves ga on. In order to generalize the data na onally, an autopsy-based,large sample size study on different age groups should be conducted.


Introduc on
Skull plays an important role, second only next to pelvis for sex determina on. In skull, mastoid and occipital region due to its anatomical loca on is the most protected region, offering high resistance to decomposi on and any form of tempering. With the intension of studying sexual dimorphism in the mastoid region involving mastoid process, we ini ated this study. The technique we adopted was first introduced by Paiva and Segre, i.e. measuring the distance between three cranio-metric land mark (Porion, Mastoidale, and Asterion) and calcula ng the area of the triangle to study sexual dimorphism.

Objec ve
The objec ve of this study is to determine sex based on mastoid triangle dimensions measured in 3D reconstructed computed tomography of the skull.

Methodology
The study included 196, 3D reconstructed CT scan images of skull, 98 of which were males and 98 of which were females. "The three craniometric points were marked to demarcate the mastoid triangle on both side". Heron's formula was used for measurement of mastoid triangle areas.

Result
Descrip ve sta s cs and the student t-test revealed that males have larger mastoid triangle dimensions in all measurements than females, with a significant difference (p<0.000) between them. The paired t-test revealed no sta s cally significant difference (p<0.05) between the right and le sides. Apart from angle asterion (p<0.05) on the right side, no other mastoid triangle angles on either side showed a significant difference. The predictability of mastoid triangle measurements and areas was determined using a discriminant func on and ROC curve analysis. The calculated mastoid triangle measurements and areas were 72.5 % of Asterion to Porion, 74.5 % of Asterion to Mastoidale, 86.3 % of Mastoidale to Porion, and 88.2 % of areas, respec vely.

Conclusion
The study findings suggest that mastoid triangle could be used to determine gender in the case of fragmentary skull remains by the forensic experts in the course of inves ga on. In order to generalize the data na onally, an autopsy-based,large sample size study on different age groups should be conducted.

INTRODUCTION
Individual iden fica on is of prime importance in mass disasters, road traffic accidents, fire accidents and in criminal cases, where there is inten onal dismembering and 1 mu la on of the body. Civilized socie es recognize the need for iden ty in living as well as in death. Iden ty of a 2 person means individuality of that person. The main objec ve of autopsy is iden fica on. Unknown bodies in full or in parts and at mes skeletal remains are brought for examina on at mortuary. Anthropological or biological iden fica on helps us in iden fica on by excluding the given popula on and also in cases where DNA and figure print cannot be obtained. Alphonso Ber lion (1853 -1914) created the first anthropometric scien fic system based on 3 physical measurements for iden fying criminals in 1880. Forensic iden fica on has evolved into an art of science 3 which involves various speciali es. Forensic anthropology is one such sub-specialty within Forensic Science, which deals 4 with human skeletonised remains and their environment. In the field of Forensic anthropology determining sex from skeletal remains, especially from isolated bones, has been an age old problem. When it comes to skeletal sex determina on, metric analysis are o en found to be of superior value as they are more objec ve but also provide greater sta s cal 5 weight than non-metric traits. Skull plays an important role in sex determina on, being the second best region for sex 4 determina on only next to pelvis. In skull the regions that 6 have been studied for sexual dimorphism are teeth, nasal 7 8 9 10 bone, frontal bone, occipital bone, foramen magnum 11 and palate. However, mastoid and occipital region due to its anatomical loca on is the most protected region, offering high resistance to decomposi on and any form of tempering.
With the intension of studying sexual dimorphism in the mastoid region involving mastoid process, we ini ated this study. The technique we adopted was first introduced by Paiva and Segre, i.e. measuring the distance between three cranio-metric land mark (Porion, Mastoidale, and Asterion) and calcula ng the area of the triangle to study sexual 12 dimorphism. However we have added other dimension to 12 the study carried out by Paiva and Segre, i.e. the angles of the mastoid triangle. Anthropological studies are subjected to intra-popula on and inter-popula on varia on; in addi on recent studies in the field of neurosurgery have ques oned 13,14 the reliability of asterion as a sta onary land mark, thus providing the ground for our study in Nepalese popula on.

METHODOLOGY
The study is a hospital-based cross-sec onal study conducted in Forensic Medicine department in collabora on with Department of Radio-diagnosis of Kathmandu University School of Medical Sciences for three month period, from December 2020 to February 2021. The sample size was 2 2 2 determined using the Cochran equa on (n = Z σ ̸ e ). A total 0 of 196 head CT scan images of adult males and females over the age of 20 years were studied. Individuals who required CT head for diagnos c purposes were randomly selected for this study. The exclusion criteria for this study were as follows: · Age less than 20 years, · Skull fractures involving the mastoid region · Pathology involving the mastoid region was not included in our study. Ins tu onal Review Commi ee of Kathmandu University School of Medical Sciences(IRC-KUSMS Approval number: 02/21), an approval was obtained in accordance with Declara on of Helsinki, in advance to ini a ng this study. 12 Paiva and Sergre had defined the anatomical landmarks of Mastoid Triangle, which became the base for this study and the landmarks were iden fied and marked on both side (right and le ) of cranium, which are as following: a  The line joining these above men oned points forms the "Mastoid Triangle". The distance between these point on both sides were measured directly on the DICOM images using inbuilt Electronic Caliper in DICOM viewer so ware in cen metres. · AP (Asterion to Porion): Distance between asterion to porion. For right side the word "R", and "L" for le was used. · MP (Mastoidale to Porion): Distance between mastoidale to porion. For right side the word "R", and "L" for le was used. · AM (Asterion to Mastoidale): Distance between asterion to mastoidale. For right side the word "R", and "L" for le was used. Original Research Ar cle Area of the both sides of mastoid triangle was calculated 2 using Heronˈs formula in cen metre square (cm ) as done by 15 Kemkes and Gobel. The angles of the mastoid triangle were calculated using online so ware Solve triangle as used 16 earlier by Kanchan et al. in which the lengths between the landmarks, asterion, porion, and mastoidale, were fed in and the angle calculated and displayed. These are documented as follows: · AngA (Angle Asterion): The angle between AM and AP. For right side the word "R", and "L" for le was used. · AngP (Angle Porion): The angle between AP and MP.
For right side the word "R", and "L" for le was used. · AngM (Angle Mastoidale): The angle between MP and AM. For right side the word "R", and "L" for le was used. Volumetric Head CT scans was done on 128 slices Simen Somatom Perspec ve Scanner at 5mm thickness with 5mm interval, the reconstruc on was done at 1mm thickness and the measurements were measured on mul -planner scan of saggital view.

Analysis
The data thus collected were entered in Microso Word in the ini al stage, which was later analysed using IBM SPSS (Sta s cal Package for Social Sciences) version 21 so ware. Descrip ve analysis was done for age and the measurements taken following with the Paired t-test done to compare right and le side of the cranium. Then Student t-test was done to equate between the sexes i.e. male and female. The level of significance is set at p-value <0.05. Discriminant Func on Analysis (DFA) and Receiver Opera ng Characteris cs (ROC) were also done.

RESULTS
Out of a total 196 individual head CT scans 98 were of males and 98 were of females. The age ranged from 20 to 89 years for males and 21 to 86 years for females, with a mean age of 44.56 (SD±17.83) years for males and 45.26 (SD±18.51) years for females (Table 1).

SD: Standard Devia on
Tables 2 shows descrip ve sta s cs and a student t-test, demonstra ng that males have larger mastoid triangle dimensions in all measurements and calcula ons. The student t-test, showed that there was a significant (p<0.000) difference between males and females in all of the measurements and calculated areas. The right and le sides of the mastoid triangle were compared using a paired t-test, which revealed no significant differences (p<0.05). However, with the excep on of angle asterion (p<0.01) on the right side, all of the angles of the mastoid triangle on both sides, i.e. right and le , showed no significant difference (p>0.05), as shown in table 3. Mastoid triangle of both right and le side has shown sexual dimorphism evidently from student t-test therefore the data was further analysed using Discriminant Func on Analysis for individual parameters, areas calculated and then all the parameters combined. Receiver Opera ng Characteris c was done for all the measurements taken and areas calculated.

C: Constant
Singh PK et al Table 6: Predictability of sex in the sample collected.  The ROC curve for all the parameters and the areas calculated of mastoid triangle for males are shown in figure  1, which showed significance in all the parameters. The highest area under the curve in males was seen in Right Area

DISCUSSION
The skull is an important structural component in determina on of sex. There are two methods for determining sex from skull measurements: morphological and morphometric. Morphometric method is more valued than morphological method as it has high sta s cal weight 15 age and is more objec ve than morphological method. Researchers have been drawn to mastoid region for sex determina on for long period. Krogman used the size of 4 mastoid process as one of his traits for sex es ma on. Similarly many studies have focused on mastoid process for sex determina on. But recent studies conducted by the neurosurgery has suggested Asterion not being a sta onary 13,14 point, which implies that Asterion is subjected to inter and intra popula on varia on. So, as Paiva and Segre suggested, this study was conducted to determine sex from mastoid triangle measurements as it is first of its kind, in Nepalese popula on. The present study has demonstrated males have larger mastoid triangle than its counterpart and there is a sta s cal significant difference (p<0.000) between them. However, there was no significant difference when right and le sides were compared. The predictability for sex es ma on was higher than in any previous study, at 73.0 % when asterion to porion of both sides was used, 70.9 % when asterion to mastoidale was used, 84.2 % when mastoidale to porion was used, and 81.1% when areas were used. When all measurements and areas calculated were obtained, the predictability increased to 95.7 %. The ROC curve further indicated the specificity and sensi vity of the various measurements and areas calculated in the mastoid triangle. A total of six studies have been carried out, in which mastoid triangle was focus of study for determina on of sex. Paiva and Segre were the first and also introduced this technique, 12 and they studied on Xerographic of the mastoid region. Our 12 study showed the same results as that of Paiva and Segre, i.e. males have larger mastoid triangle and there was significant difference between male and female. The studies that were conducted a er Paiva and Segre were done on dry skulls or at autopsy as suggested by Paiva and Segre, but our study was conducted in 3D-CT head. Similar results were also documented by other researchers in different popula on, except for study conducted by Galdames 19 in Thai popula on conducted by Manoonpolet al. Nevertheless, the predictability of sex was higher in our study in comparison to all the six studies conducted in different popula on. Therefore, our study confirms that es ma ng sex in adult human skulls is based on males' larger size and robustness in comparison to females, which is a ributed to the rate of skull growth, as male skulls experience growth even before puberty, which females do not. Features whose development is related to major muscle a achments and their ac on have been reported to be be er indicators of sex. This may be the reasons for sexual dimorphism observed in the mastoid triangle in our and other studies in different popula ons. The disparity in measurements and areas calculated in our study versus other studies could be a ributed to interpopula on differences. Therefore, our study confirms that es ma ng sex in adult human skulls is based on males' larger size and robustness in comparison to females, which is a ributed to the rate of skull growth, as male skulls experience growth even before puberty, which females do not. Features whose development is related to major muscle a achments and their ac on have been reported to be be er indicators of sex. This may be the reasons for sexual dimorphism observed in the mastoid triangle in our and other studies in different popula ons. The disparity in measurements and areas calculated in our study versus other studies could be a ributed to inter-popula on differences.

CONCLUSION
The study findings suggest that mastoid triangle could be used to determine gender in the case of fragmentary skull remains by the forensic experts in the course of inves ga on. However, further research with a larger sample size is needed to emphasize the aforemen oned points.

LIMITATIONS OF THE STUDY
Though our study demonstrates mastoid triangle as a good indicator for sex determina on, the sample size may be limita on of our study. In order to generalize the data na onally, an autopsy-based,large sample size study on different age groups should be conducted.