CBCT Analysis of Maxillary and Sphenoidal Sinuses for Forensic Age and Gender Estimation: A Retrospective Study

1. Postgraduate Student, Department of Oral Medicine and Radiology, K M Shah Dental College and Hospital, Sumandeep Vidyapeeth (Deemed to be University), Vadodara, Gujarat, India. 2. Professor and Head, Department of Oral Medicine and Radiology, K M Shah Dental College and Hospital, Sumandeep Vidyapeeth (Deemed to be University), Vadodara, Gujarat, India. 3. Professor, Department of Oral Medicine and Radiology, K M Shah Dental College and Hospital, Sumandeep Vidyapeeth (Deemed to be University), Vadodara, Gujarat, India. 4. Professor, Department of Oral Medicine and Radiology, K M Shah Dental College and Hospital, Sumandeep Vidyapeeth (Deemed to be University), Vadodara, Gujarat, India.

Abstract

Introduction: Age and gender estimation are fundamental in forensic identification, especially in crime investigation, cases of mass disasters and unidentified body remains. Paranasal sinuses, especially maxillary and sphenoidal, are unique, durable, and sexually dimorphic and well-protected within skull, resistant to trauma and decomposition. Cone Beam Computed Tomography (CBCT) enables precise Three-dimensional (3D) morphometric analysis of these anatomical structures, making them valuable markers for forensic age and gender determination.

Aim: To evaluate the morphological (volumetric) characteristics of the maxillary and sphenoidal sinuses using CBCT to assess their potential in forensic age and gender estimation.

Materials and Methods: The present Institutional-based retrospective study was conducted between September 2025 to November 2025 on CBCT archives from Department of Oral Medicine and Radiology, K M Shah Dental College and Hospital, Sumandeep Vidyapeeth (Deemed to be University), Vadodara, Gujarat, India. CBCT Digital Imaging and Communications in Medicine (DICOM) files of maxilla/both arches covering maxillary and sphenoidal sinus was selected for analysis. Total CBCT scans of 142 individuals aged 21-75 years were analysed and volumetric measurements of bilateral maxillary and sphenoidal sinus were recorded by semi-automatic segmentation method using 3D Slicer software and statistically analysed to determine correlations with age and gender. Inclusion and exclusion criteria were strictly applied to ensure sample consistency. For statistical analysis Chi-square test, Mann-Whitney, Kruskal-Wallis tests and Logistic regression analysis were done, p-values less than 0.05 were regarded as statistically significant.

Results: Significant variations in sinus dimensions were observed between different genders and age groups. The median volumes of the maxillary sinus (males: 13,574.75 mm³; females: 11,210.30 mm³) and the sphenoidal sinus (males: 4,918.71 mm³; females: 4,017.23 mm³) were significantly greater in males than in females, with p-values of 0.023 and 0.02, respectively. Correlation analysis revealed a statistically significant (p-value <0.05) negative association between age and sinus volume, with advancing age linked to reduced volumes in the bilateral maxillary sinuses and the right sphenoidal sinus.

Conclusion: The CBCT imaging of the maxillary and sphenoidal sinuses provides valuable morphometric data that can aid in forensic identification. These anatomical structures demonstrate measurable differences across age and gender, supporting their utility as supplementary tools in forensic profiling.

Cone beam computed tomography, Image processing, Paranasal sinus, Three-dimensional imaging, Volumetric analysis

The primary focus of forensic radiological evaluation is the osseous skeleton, which plays a vital role in forensic medicine, particularly in anthropometric assessments. Bones commonly utilised for forensic identification such as the pelvis, skull, and long bones are often recovered in fragmented, incomplete, or commingled conditions, especially in mass disaster scenarios like airplane crashes, explosions, armed conflicts, and natural calamities. These circumstances pose significant challenges to both identification and accurate personal identification (1),(2). The craniofacial skeleton exhibits distinct sexual dimorphism and age-related morphological changes, making it a valuable tool in forensic dentistry for identification purposes. With ongoing advancements in diagnostic imaging, traditional two-dimensional radiography is increasingly being supplanted by three-dimensional modalities such as Computed Tomography (CT) and CBCT, offering enhanced accuracy and detail in forensic investigations (3)[,4]. Evaluation of the paranasal sinuses for forensic purposes has proven to be a reliable method. Several studies have indicated that the jaw and skull bones often remain intact, even in cases involving trauma or decomposition, making them valuable structures for identification (5),(6),(7). The paranasal sinuses are air-filled chambers found in the bones that encircle the nasal cavities. The maxillary sinuses, located bilaterally within the maxillary bones, are the largest and earliest to develop among the paranasal sinuses. In contrast, the sphenoid sinuses, deeply embedded within the sphenoid body, benefit from a well-protected anatomical position that offers substantial resistance to externally induced traumatic damage (5),(6),(7).

The CBCT has emerged as a valuable imaging modality in oral and maxillofacial applications, offering three-dimensional visualisation similar to conventional CT but with several distinct advantages. Compared to traditional two-dimensional imaging, CBCT provides enhanced spatial accuracy and depth. Notably, it surpasses conventional CT in terms of lower radiation exposure, cost-effectiveness, wider availability, reduced scanning time, high-resolution output, and ease of image interpretation. CBCT data is stored in the standardised DICOM format, facilitating seamless telecommunication and compatibility with third-party imaging software. This format also allows for selective voxel display within the dataset, enabling precise volumetric visualisation (8).

Many medical image annotation software like Insight Segmentation and Registration Toolkit (ITK)-SNAP, 3D Slicer,Materialise Interactive Medical Image Control System (MIMICS) etc., supports DICOM files and has segmentation tools for 3D volumetric analysis of anatomy and pathology. The 3D Slicer is available for free download and can be used on all platforms (Windows, macOS, Linux) and is completely open source. 3D Slicer software has AI-based automatic segmentation feature and also capable of supporting hundreds of segments per image [9,10]. Previous studies have demonstrated that sexual dimorphism exists in the anthropometric measurements of maxillary sinuses and sphenoidal sinus. However, mostly the measurements like height, width and length of the sinuses are considered. Very few researchers have done the volumetric analysis of the sinuses for forensic evaluation and many of the studies are CT based studies (11),(12),(13),(14),. Therefore, the present study was designed with an aim to evaluate the volumetric morphology of the maxillary and sphenoidal sinuses using CBCT, with the objective of assessing their potential utility in age and gender estimation.

The null hypothesis (H0) was that there is no significant difference in the volumetric measurements of the maxillary and sphenoidal sinuses between males and females, or among different age groups, as measured by CBCT. Furthermore, there is no significant correlation between sinus volumes and demographic variables (gender and age).

The alternative hypothesis (H1) was that there is a significant difference in the volumetric measurements of the maxillary and sphenoidal sinuses between genders and/or among different age groups, as measured by CBCT. Additionally, there exists a significant correlation between sinus volumes and demographic variables (gender and age).

Material and Methods

The present Institutional-based retrospective study was conducted between September 2025 to November 2025 on CBCT archives from Department of Oral Medicine and Radiology, K M Shah Dental College and Hospital, Sumandeep Vidyapeeth (Deemed to be University), Vadodara, Gujarat, India. CBCT DICOM files of maxilla/both arches covering maxillary and sphenoidal sinus taken between September 2024 to October 2025 were selected from archives for analysis. The study was approved by Institutional Ethical Committee (SVIEC/ON/DENT/SRP/Sep- 25/20 dated 08/09/2025).

Sample size calculation: Sample size calculation was based on the findings of the study conducted by Sivasamy I et al., (7). In this article the right maxillary sinus volume showed a significant difference between males (mean=16147.98±5684.88 mm³) and females (mean=12973.76±3904.85 mm³), with a mean difference of 3174.22 mm³. Using these values, the pooled standard deviation was calculated as 4870.9 mm³, yielding an effect size (Cohen’s d) of 0.65. Based on a two-sample t-test with a significance level of 0.05 and a power of 80%, the minimum required sample size was estimated to be 39 participants per group (approximately 78 in total). However, to enhance the robustness and generalisability of the findings, 142 CBCT scans were included in the study, thereby ensuring higher statistical power and reliability of results.

Inclusion and Exclusion criteria: CBCT DICOM files of maxillary arch (FOV: 12X5) covering maxillary and sphenoidal sinus of individuals between 21 to 75 years irrespective of gender were included for the study. The images were acquired from CS 9600 CBCT unit with exposure parameters of 120 KVp, 6.3 mA and 20.0 sec exposure time.CBCT files with poor quality image, artifacts and any pathology in maxillary and sphenoidal sinus were excluded.

Study Procedure

Total 320 CBCT scans were screened, verified for complete maxillary and sphenoidal sinus visibility and 142 scans matching the inclusion criteria were selected for further evaluation. The selected scans were exported to 3D Slicer (5.8.1) application for volumetric analysis. 3D Slicer was selected over other open-source software as it supports multidimensional data, allows measurement of areas, distances, angles, and volumetric calculations. Despite its advanced capabilities 3D Slicer has a user friendly interface [9,10]. Bilateral maxillary and sphenoidal sinuses were identified and independently each sinus volumetric analysis was performed using semi-automatic segmentation method.The intraclass correlation coefficient analysis demonstrated excellent inter observer (0.9) reliability for all sinus volume measurements.

The patient’s DICOM dataset was imported into 3D Slicer and processed within the segment editor module. A new segmentation was created, and the paranasal sinus regions were isolated using the thresholding tool, which allowed differentiation of air filled cavities from surrounding osseous and soft-tissue structures. The intensity range was carefully adjusted to exclude cortical bone and mucosal lining, ensuring that only the aerated sinus volume was captured. To refine the segmentation, smoothing filters (such as median or Gaussian smoothing) were applied, which minimised irregular boundaries and enhanced anatomical fidelity.

Following segmentation, the segment statistics module was employed to compute volumetric measurements. This tool automatically calculates the three-dimensional volume of the segmented region, expressed in cubic millimetres, providing reproducible quantitative data. The segmented sinus was then visualised in the 3D rendering view, enabling verification of anatomical accuracy and correction of any segmentation errors. Finally, the volumetric results and 3D models were exported for documentation, statistical analysis, and integration into the research dataset (Table/Fig 1).

STATISTICAL ANALYSIS

Differences in sample size between genders across age groups were tested using the chi-square test. All the parameters were checked for their normality using Shapiro-Wilk test. It was found that all variables were violating normality assumptions and hence further analysis was carried out using “Non-parametric” (Mann-Whitney, Kruskal-Wallis) tests. Logistic regression analysis was done to check the accuracy of paranasal sinuses for gender determination. Standard deviations and means were used to express the variables. The p-values less than 0.05 were regarded as statistically significant.

Results

Total 142 (n) scans out of the 320 CBCT volumes were selected as per the inclusion and exclusion criteria for 3D volumetric assessment of maxillary and sphenoidal sinus. Out of 142 scans 70 (49.3%) scans were from males and 72 (50.7%) scans were from females. Further the scans were divided in to five sub-groups based on age. The male to female distribution was not significant amongst different age groups (Chi-square value 2.713; p-value 0.607) (Table/Fig 2).

The total median volume of the maxillary was significantly greater in males (13,574.75 mm³) compared to females (11,210.30 mm³), with the difference reaching statistical significance (p=0.023). Sidewise median volumetric measurements in males were found to be 14,640.80 mm³ on the right side and 12,472.50 mm³ on the left. In females, the right maxillary sinus volume measured 11,994.50 mm³, while the left-side measured 10,999.60 mm³. The sphenoidal sinus volume was also greater in males (4,918.71 mm³) than in females (4,017.23 mm³), with the difference being statistically significant. With respect to sphenoidal sinus, sidewise volumetric measurements in males were found to be 5403.22 mm³ on the right-side and 5102.51 mm³ on the left. In females, the right sphenoidal sinus volume measured 3650.32 mm³, while the left side measured 4315.01 mm³. Further analysis revealed that the statistically significant difference in sinus volume between genders was confined to the right side for both the maxillary and sphenoidal sinuses (Table/Fig 3), (Table/Fig 4).

Gender estimation accuracy was assessed using logistic regression analysis, yielding a combined accuracy of 62% for bilateral maxillary and sphenoidal sinus volumes and the results are detailed in (Table/Fig 5). Age-related changes in sinus volume were statistically evaluated using the Kruskal-Wallis test. The mean volumes of both the maxillary and sphenoidal sinuses demonstrated a decreasing trend with advancing age. Statistically significant differences were observed in the bilateral maxillary sinuses and the right sphenoidal sinus (Table/Fig 6). Spearman’s rank correlation analysis revealed a strong negative correlation for both the right and left maxillary sinuses with age, with p-values less than 0.001 (Table/Fig 7).

Discussion

Paranasal sinuses, as unique anatomical structures, exhibit considerable individual variability that remains relatively preserved even after skeletal decomposition. Consequently, sinus morphometry provides valuable adjunctive information for forensic identification, especially when other skeletal markers are fragmented or unavailable (1),(3),(4). The present retrospective study aimed to evaluate the maxillary and sphenoidal sinuses volume using CBCT to assess their potential role in forensic age and gender estimation. In this study, CBCT DICOM images were used for volumetric assessment of the maxillary and sphenoidal sinuses through segmentation methods in 3D Slicer. Researchers have also employed ITK-SNAP, MIMICS, and other commercial real-time segmentation software for volumetric analysis of various anatomical structures (5),(7),(12).

In the present study, the volumes of both the maxillary and sphenoidal sinuses were found to be significantly greater in males than in females, thereby leading to rejection of the null hypothesis.These findings are consistent with those of previous morphometric studies-based on two-dimensional measurements and/or three-dimensional analyses-conducted using various imaging modalities such as lateral cephalograms, CT scans, and CBCT. Sivasamy I et al., conducted a study on a central Indian population, reporting average maxillary sinus volumes of 15,526.65 mm³ in males and 12,980.06 mm³ in females. The sphenoidal sinus volumes were 4,380.22 mm³ in males and 4,054.08 mm³ in females. These measurements closely align with those observed in our study of a western Indian population, reinforcing the utility of both sinuses in forensic sex determination (7). The binary logistic regression accuracy was modest, with an overall correct prediction rate of 62%. Similar accuracy rates were shown in a study conducted by Sivasamy I et al., (7). Research conducted by Vidya CS et al., examined 30 buried skulls of known sex, which were imaged using 3D axial multislice CT scans. Focusing on the maxillary sinuses, the researchers used SyngoVolume (Siemens) software to automatically estimate the bilateral maxillary antral volumes. The study found that the maxillary sinus volume was greater in males compared to females (11). Uthman AT et al., used helical CT for measuring the width, length, height, and total distance across both maxillary sinuses. The authors found that the maxillary sinus height showed sexual dimorphism, with an accuracy rate of 71.6% (15). Similarly, Gibelli D et al., reported mean sphenoid sinus volumes of 10.005±5.101 cm³ in males and 7.920±3.176 cm³ in females in their volumetric assessment study (16). Waluyo RF et al., conducted a study using CBCT images that revealed sexual dimorphism in the two-dimensional parameters of the maxillary sinus (17). Present study revealed the accuracy of gender determination for as 60.6% and 57.7% for right and left maxillary sinus respectively. Sidhu R et al., measured the dimensions of the maxillary sinus using lateral cephalograms, and their Discriminant Function (DF) analysis correctly classified 76% of the original grouped cases, with an overall sensitivity of 80% and specificity of 72% (18). Furthermore, a DF analysis based on the sinus volumes and dimensions achieved a gender prediction accuracy of 69.81% in a study conducted by Sharma SK et al., (19). In contrast, Barros FD et al., did not observe significant differences in maxillary sinus area or volume between sexes (20).

In the present study, adult sinus volumes were categorised based on different age groups. The mean volumes of both the maxillary and sphenoidal sinuses showed a significant reduction with advancing age, indicating a negative correlation between age and sinus volume and thereby supporting rejection of the null hypothesis. This feature may be useful in forensic age estimation. Similarly, Ariji Y et al., in their CT-based study, observed that maxillary sinus volume increased up to the age of 20 years, followed by a gradual decline (14). In forensic odontology, paranasal sinus measurements have proven to be reliable indicators for age and gender identification. The findings of the present study demonstrated that CBCT-derived volumetric data can be effectively utilised for the analysis of paranasal sinuses, thereby supporting their application in forensic identification.

Limitation(s)

As the present study was conducted on a population from western India (Gujarat), the findings may not be generalisable to other regions. To develop a comprehensive database for forensic investigations, further research employing a multicentre approach across diverse geographic populations is recommended.

Conclusion

The CBCT of the maxillary and sphenoidal sinuses yields critical morphometric data useful for forensic age identification and also has 62% accuracy for gender determination. These anatomical features exhibit quantifiable variations with respect to age and sex and hence it can be used along with other craniofacial parameters in forensic profiling.

DOI and Others

DOI: 10.7860/JCDR/2026/85630.24372

Date of Submission: Nov 29, 2025
Date of Peer Review: Jan 20, 2026
Date of Acceptance: May 20, 2026
Date of Publishing: Sep 01, 2026

AUTHOR DECLARATION:
• Financial or Other Competing Interests: None
• Was Ethics Committee Approval obtained for this study? Yes
• Was informed consent obtained from the subjects involved in the study? No
• For any images presented appropriate consent has been obtained from the subjects. No

PLAGIARISM CHECKING METHODS:
• Plagiarism X-checker: Dec 10, 2025
• Manual Googling: May 16, 2026
• iThenticate Software: May 18, 2026 (9%)

ETYMOLOGY: Author Origin

EMENDATIONS: 9