Morphometric analysis of the skull base and palatal regions for gender identification using CBCT: a retrospective study

Introduction

Forensic identification is a crucial aspect of forensic science which uses scientific techniques to identify unknown individuals or provide evidence for criminal investigations (Issrani et al., 2022). Gender determination is one area of focus in forensic identification, as it can provide valuable information in the identification process (Nagare et al., 2018).

Cone beam computed tomography (CBCT) imaging has emerged as a powerful and superior tool in the subjective evaluation of the craniofacial complex. This advanced imaging technology provides detailed three-dimensional images of the dental and craniofacial structures, offering significant advantages over traditional two-dimensional radiography methods.

One of the key strengths of CBCT lies in its ability to capture high-resolution images with minimal distortion. Traditional radiographic techniques, such as panoramic and cephalometric imaging, often suffer from overlapping structures, leading to a lack of clarity and precision. In contrast, CBCT produces clear, detailed images that allow for a comprehensive assessment of the entire craniofacial region. The three-dimensional visualization provided by CBCT enables accurate linear measurement of craniofacial lines and angles, aiding in forensic identification (Ajmal et al., 2023; Shaweesh et al., 2023). CBCT provides highly detailed 3D images of the cranium, allowing for precise measurements of various cranial characteristics that can be used to determine gender (Moreira et al., 2009). CBCT offers several advantages over traditional methods such as visual inspection or manual measurements, as it provides more precise and objective measurements and is less susceptible to error and subjective interpretation (Aksoy et al., 2016). Additionally, CBCT can provide information about a person’s ethnicity, age, and ancestry, which can be beneficial for forensic investigations (Zamani Naser & Panahi Boroujeni, 2014). Radiography, including CBCT scans, is commonly used to compare the palatal width and height in forensic identification (Magat & Ozcan, 2019). CBCT can also be used to evaluate abnormalities of the skull base region, which can provide crucial information for forensic investigations (Marques et al., 2012). As a person matures, the shape and size of the skull base region change, and these changes can be observed in CBCT images (Magat & Ozcan, 2019). This information can be useful in cases involving unidentified remains where the age of the person is unknown.

The purpose of this study was to evaluate the accuracy of morphometry of palatal and skull base region in gender discrimination using CBCT scanning and to assess the accuracy of the results among a sample of the Arab population. The research rationale is to provide vital insights into the usefulness and precision of CBCT in forensic science by examining these anatomical landmarks within a specific demography, such as the Arab community. Improving the accuracy of gender identification using CBCT can greatly enhance forensic investigations, helping to solve instances involving unidentified individuals and making valuable contributions to the wider area of forensic anthropology and science.

Patients & Methods

The radiographic study analysis included 142 CBCT scans of consented patients who underwent different dental operations at the University Dental Hospital, Sharjah (UDHS). Each participant provided written informed consent in accordance with the specific CBCT procedure. This study followed all protocols laid out in the 2014 Declaration of Helsinki and had the approval of the University of Sharjah’s Human Subjects Research Ethics Committee (REC-19-01-24-01). We determined an a priori sample size of 140 using the G power analysis software, with the following parameters: effect size = 0.5, error probability = 0.05, and study power = 0.9. The imaging settings used for the scans were as follows: a 17 ×13 cm field of view, 85 kVp, 7 mA, 14 s exposure, and a voxel size of 0.25 mm. The scanner was Galileo’s Comfort CBCT from Sirona Dental Systems in Bensheim, Germany. The CBCT images were displayed on a 23-inch DELL monitor with 1920 ×1080 pixels using the SIDEXIS program. The CBCT pictures were reviewed by two experienced dental radiologists (AU and SS), and a third expert (NA) was brought in to help out if the two of them couldn’t come to a consensus. These differences could be due to poor picture quality, differences in the chosen CBCT slice used to determine distances, or differences in the recognition of anatomical landmarks.

Definition of anatomical landmarks used

Skull base parameters

Transverse diameter of foramen magnum (TDFM): The maximum horizontal distance of foramen magnum on axial plane. Marques et al. (2012) (Fig. 1, point 1).

Width of right to left occipital condyle (WOC): The greatest distance from the left occipital condyle to the right occipital condyle on axial plane (Abdel-Karim, Housseini & Hashish, 2015). (Fig. 1, point 2).

The occipital condyle height (right and left) (OCH) was measured on coronal CT images as the length of a line centered in the occipital condyle between the midpoint of its superior articular surface and the midpoint of its inferior articular surface (Abdel-Karim, Housseini & Hashish, 2015) (Fig. 1, point 3).

On a midline sagittal image, the height of the crista galli (HCG) and the distance between the crista galli and anterior tuberculum sellae (CG-ATS) were measured.

HCG was defined as the maximum distance between the apex and base of the crista galli. (Marques et al., 2012) (Fig. 2, point 1).

The CG-ATS was determined by measuring the distance from the back edge of the crista galli to the area of the anterior tuberculum sellae, extending to the end of the planum sphenoidale (Batista et al., 2015) (Fig. 2, point 2).

In order to find the basal angle (BA), two lines were drawn on the sagittal plane, one from the nasion to the dorsum sellae and the other from the dorsum sellae to the basion (Botelho & Ferreira, 2013) (Fig. 2, point 3).

The clivus-canal angle (CCA) is determined by measuring the angle formed by a line extending from the lower one-third of the clivus and a line extending from the lower back part of the C-2 vertebral body to the upper back portion of the dens on the sagittal plane. (Botelho & Ferreira, 2013) (Fig. 2, point 4).

Clivus length (CL) was determined by calculating the distance between the dorsum sellae and the bason on a midline sagittal plane (Dufton et al., 2011) (Fig. 2, point 5).

Palatal width (PW): The distance between the cemento-enamel junction of the left and right first molars from coronal plane (Spradley & Jantz, 2011) (Fig. 3, point 1).

Palatal height (PH): The distance of the perpendicular line drawn from midpalatal region intersecting the horizontal line connecting the cemento-enamel junction of the left and right first molars from coronal plane. (Spradley & Jantz, 2011) (Fig. 3, point 2).

Results

Discussion

The skull base is crucial in forensic identification due to its unique anatomical features that aid in determining various aspects of an individual’s identity. Gender identification is facilitated by sexual dimorphism in the skull base, with studies showing that discriminant analysis of skull base dimensions can accurately determine an individual’s sex (Spradley & Jantz, 2011; Walker, 2008). This underscores the skull base’s importance in gender identification within forensic anthropology. The present study support this notion were the clivus length, occipital condyle height, basal angle and transverse diameter of foramen magnum were among the other cranium base region parameters that were significantly greater in males than in females. Additionally, the skull base is essential for establishing the biological profile of an individual, including factors such as age, sex, ancestry, and stature. Combined with other skeletal elements, the skull base provides valuable information for comprehensive individual identification (Baryah, Krishan & Kanchan, 2019). Moreover, the skull base is vital for facial reconstruction, serving as the foundation for reconstructing soft tissue features and aiding in visual representation for identification purposes (Bonda, 2018). Furthermore, the skull base is utilized in ancestry estimation, where its features are assessed to determine an individual’s ancestral background. This information is crucial in narrowing down potential matches in missing persons cases or unidentified remains (Pengyue et al., 2021). Advanced techniques like deep neural networks can be employed to analyze the skull base’s morphological characteristics for insights into an individual’s ancestry (Pengyue et al., 2021). Palatal width and height were assessed by Ajanovic et al. (2022), who used the method of geometric morphometry and found that the determination of sex based on the size and shape of the hard palate was possible with an accuracy of 58.57% for women and 66.91% for men in the sample from the population of Bosnia and Herzegovina. The accuracy of their study was slightly lower than that observed in the present study (60%) for women and moderately higher than that of our study (59.7%) for men. In the study conducted by Uthman, Al-Rawi & Al-Timimi (2012), the dimensions of the foramen magnum were examined to determine differences between genders. The researchers found that the circumference and area of the foramen magnum were the most effective parameters for studying sexual dimorphism, with an accuracy of 67% and 69.3% respectively. These findings are similar to the results of the current study, which showed an accuracy of 61.1% for men and 57.1% for women. Through the application of multivariate analysis, it was determined that 90.7% of the FM dimensions of males and 73.3% of the FM dimensions of females were accurately classified by sex (Uthman, Al-Rawi & Al-Timimi, 2012). The height of the occipital condyle has been specifically studied in relation to gender identification in relation to morphological analysis of the foramen magnum (Chethan et al., 2012). They found that the shape of the foramen magnum varied among individuals, and this information could be useful for anthropologists, morphologists, and clinical anatomists. Abo El-Atta et al. (2020) examined the sexual dimorphism of the occipital condyles and foramen magnum in an Egyptian population. Males and females differed significantly in ten dimensions of the occipital condyles and foramen magnum (Abo El-Atta et al., 2020).

The dimensions and characteristics of various anatomical structures, such as the frontal sinuses, maxillary sinuses, and mastoid process, have been analyzed using discriminant analysis to determine their potential for gender determination. Discriminant analysis methods have proven to be effective in accurately categorizing individuals based on their gender. Jayakrishnan, Reddy & Kumar (2023) conducted a study using discriminant function analysis and found that 91.3% of females and 50% of males correctly identified their gender based on paranasal sinus dimensions. These findings demonstrate the potential of discriminant analysis methods in CBCT for gender identification. Recently researchers have also found success in gender determination using parameters like intra mastoid diameter and inter zygomatic width (Jayakrishnan, Reddy & Kumar, 2023; Albalawi et al., 2019). Some of the recent studies have also investigated the CBCT variables in the mandible for possible role in gender determination (Rathithya et al., 2023). Rathithya et al. (2023) found that the distance between the anterior and posterior mandibular foramen is good predictor for gender determination. Another CBCT based study by Albalawi et al. (2019) reported that angle formed by the intersection of lines from the left and right gonion to mention could be useful in gender determination. However, another recent study by Bakan et al. (2022) revealed that CBCT based gonial angle measurements were not very precise in determining gender.

The future scope of work in this field would depend on the application of machine learning algorithms. Recent studies have revealed that the Gaussian naive Bayes machine learning model showed high accuracy (90%) in gender determination using CBCT (Baban & Mohammad, 2023).

Conclusion

The use of discriminant analysis in CBCT for gender identification offers several advantages. It provides an efficient method for determining gender, which is particularly valuable in forensic science and anthropological research. Discriminant analysis related to gender showed that occipital condyle height to be the most accurate and best discriminator among the skull base region parameters (AUC = 0.722). The overall accuracy for gender discrimination using all parameter was 77.9%. Palatal dimensions yielded accuracy in females’ identification by 60%. By analyzing the dimensions and characteristics of specific anatomical structures, discriminant analysis can provide reliable and accurate results for gender determination.

Supplemental Information