Different machine learning methods based on maxillary sinus in sex estimation for northwestern Chinese Han population.

BACKGROUND & OBJECTIVE: Sex estimation is a critical aspect of forensic expertise. Some special anatomical structures, such as the maxillary sinus, can still maintain integrity in harsh environmental conditions and may be served as a basis for sex estimation. Due to the complex nature of sex estimation, several studies have been conducted using different machine learning algorithms to improve the accuracy of sex prediction from anatomical measurements. MATERIAL & METHODS: In this study, linear data of the maxillary sinus in the population of northwest China by using Cone-Beam Computed Tomography (CBCT) were collected and utilized to develop logistic, K-Nearest Neighbor (KNN), Support Vector Machine (SVM) and random forest (RF) models for sex estimation with R 4.3.1. CBCT images from 477 samples of Han population (75 males and 81 females, aged 5-17 years; 162 males and 159 females, aged 18-72) were used to establish and verify the model. Length (MSL), width (MSW), height (MSH) of both the left and right maxillary sinuses and distance of lateral wall between two maxillary sinuses (distance) were measured. 80% of the data were randomly picked as the training set and others were testing set. Besides, these samples were grouped by age bracket and fitted models as an attempt. RESULTS: Overall, the accuracy of the sex estimation for individuals over 18 years old on the testing set was 77.78%, with a slightly higher accuracy rate for males at 78.12% compared to females at 77.42%. However, accuracy of sex estimation for individuals under 18 was challenging. In comparison to logistic, KNN and SVM, RF exhibited higher accuracy rates. Moreover, incorporating age as a variable improved the accuracy of sex estimation, particularly in the 18-27 age group, where the accuracy rate increased to 88.46%. Meanwhile, all variables showed a linear correlation with age. CONCLUSION: The linear measurements of the maxillary sinus could be a valuable tool for sex estimation in individuals aged 18 and over. A robust RF model has been developed for sex estimation within the Han population residing in the northwestern region of China. The accuracy of sex estimation could be higher when age is used as a predictive variable.

Sex estimation using maxillary sinus volume for Chinese subjects based on cone-beam computed tomography.

BACKGROUND: Sex estimate is a key stage in forensic science for identifying individuals. Some anatomical structures may be useful for sex estimation since they retain their integrity even after highly severe events. However, few studies are focusing on the Chinese population. Some researchers used teeth for sex estimation, but comparison with maxillary sinus were lack. As a result, the objective of this research is to develop a sex estimation formula for the northwestern Chinese population by the volume of the maxillary sinus and compare with the accuracy of sex estimation based on teeth through cone-beam computed tomography (CBCT). METHODS: CBCT images from 349 samples were used to establish and verify the formula. The volume of both the left and right maxillary sinuses was measured and examined for appropriate formula coefficients. To create the formula, we randomly picked 80% of the data as the training set and 20% of the samples as the testing set. Another set of samples, including 20 males and 20 females, were used to compare the accuracy of maxillary sinuses and teeth. RESULTS: Overall, sex estimation accuracy by volume of the left maxillary sinus can reach 78.57%, while by the volume of the right maxillary sinus can reach 74.29%. The accuracy for females, which can reach 91.43% using the left maxillary sinus, was significantly higher than that for males, which was 65.71%. The result also shows that maxillary sinus volume was higher in males. The comparison with the available results using measurements of teeth for sex estimation performed by our group showed that the accuracy of sex estimation using canines volume was higher than the one using maxillary sinus volume, the accuracies based on mesiodistal diameter of canine and first molar were the same or lower than the volume of maxillary sinus. CONCLUSIONS: The study demonstrates that measurement of maxillary sinus volume based on CBCT scans was an available and alternative method for sex estimation. And we established a method to accurately assess the sex of the northwest Chinese population. The comparison with the results of teeth measurements made the conclusion more reliable.

Dental Age Estimation in Northern Chinese Han Children and Adolescents Using Demirjian's Method Combined with Machine Learning Algorithms.

OBJECTIVES: To investigate the application value of combining the Demirjian's method with machine learning algorithms for dental age estimation in northern Chinese Han children and adolescents. METHODS: Oral panoramic images of 10 256 Han individuals aged 5 to 24 years in northern China were collected. The development of eight permanent teeth in the left mandibular was classified into different stages using the Demirjian's method. Various machine learning algorithms, including support vector regression (SVR), gradient boosting regression (GBR), linear regression (LR), random forest regression (RFR), and decision tree regression (DTR) were employed. Age estimation models were constructed based on total, female, and male samples respectively using these algorithms. The fitting performance of different machine learning algorithms in these three groups was evaluated. RESULTS: SVR demonstrated superior estimation efficiency among all machine learning models in both total and female samples, while GBR showed the best performance in male samples. The mean absolute error (MAE) of the optimal age estimation model was 1.246 3, 1.281 8 and 1.153 8 years in the total, female and male samples, respectively. The optimal age estimation model exhibited varying levels of accuracy across different age ranges, which provided relatively accurate age estimations in individuals under 18 years old. CONCLUSIONS: The machine learning model developed in this study exhibits good age estimation efficiency in northern Chinese Han children and adolescents. However, its performance is not ideal when applied to adult population. To improve the accuracy in age estimation, the other variables can be considered.

Haplogroup Structure and Genetic Variation Analyses of 60 Mitochondrial DNA Markers in Southern Shaanxi Han Population.

Mitochondrial DNA (mtDNA) has been widely employed as one tool for the studies of human migration and phylogenetic evolution owing to the characteristics of its lack of recombination and matrilineal inheritance. In this study, we analyze genetic distributions of 60 mtDNA markers in 126 unrelated individuals of Southern Shaanxi Han population and classify their haplogroups. Genetic distribution comparisons between Southern Shaanxi Han and other populations from different continents are conducted based on the same mtDNA markers. The majority of 60 mtDNA markers are polymorphic in Southern Shaanxi Han population. The most common haplogroups observed in Southern Shaanxi Han population are B5, followed by D5, A, D4e, and N9a1'3. Obtained matching probability for these 60 mtDNA markers indicates that the panel could be used as a valuable tool in forensic caseworks. Results of genetic distances (Fst) and multidimensional scaling analysis show that Southern Shaanxi Han population has relatively close genetic relationships with other Han populations in different regions. In conclusion, the panel comprising 60 mtDNA markers could be utilized for forensic applications in Southern Shaanxi Han population.

Genetic distribution analyses and population background explorations of Gansu Yugur and Guizhou Miao groups via InDel markers.

Insertion and deletion markers (InDels) have gained considerable attentions in population genetics and forensic research. In this study, we investigated genetic distributions of 30 InDels in Gansu Yugur and Guizhou Miao groups and evaluated their forensic application values. Genetic relationship analyses between Gansu Yugur, Guizhou Miao groups and other published populations were conducted based on these 30 InDels. Power of discrimination and power of exclusion in trio and duo cases of 30 InDels ranged from 0.3528 to 0.6247, 0.0937 to 0.1873, and 0.0219 to 0.1247 in Gansu Yugur group; and they ranged from 0.2579 to 0.6247, 0.0671 to 0.1874, and 0.0105 to 0.1247 in Guizhou Miao group. Obtained cumulative power of discrimination values indicated these InDels could be used for forensic individual identifications in both ethnic groups. Principal component analysis and phylogenetic reconstruction revealed that Gansu Yugur and Guizhou Miao groups had close affinities with their neighboring populations. Genetic structure analyses among these populations also indicated that studied Gansu Yugur and Guizhou Miao groups showed similar genetic structure with their neighboring populations. Further analyses of Y-STR, mtDNA, and ancestry informative markers should be conducted to better understand genetic backgrounds of Gansu Yugur and Guizhou Miao groups in the future.

Chinese Xibe population genetic composition according to linkage groups of X-chromosomal STRs: population genetic variability and interpopulation comparisons.

BACKGROUND: The Xibe population is one of China's officially recognised populations and is now distributed separately from west to east in the northern part of China. X-chromosomal short tandem repeats have a special inheritance pattern, and could be used as complements in forensic application, especially for complex or deficiency cases. SUBJECTS AND METHODS: This study obtained the allelic and haplotypic frequencies of 19 X-STR loci in the Xibe population from Xinjiang Uygur Autonomous Region, China, and studied the genetic differentiations between the Xibe and other populations. RESULTS: The combined power of discrimination in females and males and mean exclusion chances in deficiency cases, normal trios and duo cases was at least 0.999 999 994. In the haplotypic study, the Xibe population showed a more similar pattern of haplotype distribution with Asian populations than populations from other continents, while allelic study also indicated a closer relationship between the Xibe and Asian populations. CONCLUSIONS: The 19 X-STR loci would be useful in forensic application in the studied population. The Xibe population showed a closer genetic relationship with Asian populations in the study, and more population data would be necessary for more detailed genetic relationship studies.

Allele and haplotype diversity of new multiplex of 19 ChrX-STR loci in Han population from Guanzhong region (China).

X-chromosomal short tandem repeats (X-STRs) have been proved to be useful for some deficiency paternity cases in recent years. Here, we studied the genetic polymorphisms of 19 X-STR loci (DXS10148-DXS10135-DXS8378, DXS10159-DXS10162-DXS10164, DXS7132-DXS10079-DXS10074-DXS10075, DXS6809-DXS6789, DXS7424-DXS101, DXS10103-HPRTB-DXS10101 and DXS7423-DXS10134) in 252 male and 222 female individuals from Guanzhong Han population, China. No deviation for all 19 loci was observed from the Hardy-Weinberg equilibrium. The polymorphism information content values of the panel of 19 loci were more than 0.5 with the exception of the locus DXS7423. The combined power of discrimination were 0.9999999999999999999994340 in females and 0.9999999999997662 in males, respectively; and the combined mean exclusion chances were 0.999999993764 in duos and 0.999999999997444 in trios, respectively. The haplotype diversities for all the seven clusters of linked loci were more than 0.9. The results showed that the panel of 19 X-STR loci were powerful for forensic applications in Guanzhong Han population. Locus by locus population comparisons showed significant differences at more than seven loci between Guanzhong Han population and the groups from North America, Europe and Africa.

24 Y-chromosomal STR haplotypic structure for Chinese Kazak ethnic group and its genetic relationships with other groups.

The Kazak ethnic minority is a large ethnic group in the Xinjiang Uygur Autonomous Region of China and is valuable resource for the study of ethnogeny. In the present study, 24 Y-chromosomal short tandem repeat (Y-STR) loci were analyzed in 201 unrelated Kazak male individuals from Ili Kazak Autonomous Prefecture, Xinjiang, China. The gene diversity of the 24 Y-STR loci in the studied Kazak group ranged from 0.0050 to 0.9104. According to haplotypic analysis of the 24 Y-STR loci, 113 different haplotypes were obtained, 96 of which were unique. The haplotype diversity and discrimination capacity in Kazak group were 0.9578 and 0.5622 at 24 STR loci, respectively. The haplotype diversity and discrimination capacity at Y-filer 17 loci, extended 11 loci, and minimal 9 loci were reduced to 0.9274 and 0.4279, 0.8459 and 0.3284, and 0.8354 and 0.2985, respectively, which could indicate that the more loci were detected, the higher forensic efficacy was obtained. We evaluated the application value of the 24 loci in forensic sciences and analyzed interpopulation differentiations by making comparisons between the Kazak1 (represent our samples from Ili Kazak Autonomous Prefecture) group and other 14 groups. The results of pairwise genetic distances, multidimensional scaling plot, and neighbor-joining tree at the same set of 17 Y-filer loci indicated that the Kazak1 group had the closer genetic relationships with Kazak2 (represent samples from the whole territory of Xinjiang Uygur Autonomous Region), Mongolian, and Uygur ethnic groups. The present results may provide useful information for paternal lineages in forensic cases and can also increase our understanding of the genetic relationships between Kazak1 and other groups.

24 Y-chromosomal STR haplotypic polymorphisms for Chinese Uygur ethnic group and its phylogenic analysis with other Chinese groups.

The Uygur ethnic minority is the largest ethnic group in the Xinjiang Uygur Autonomous Region of China, and is a precious resource for the study of ethnogeny and forensic biology. Previous studies have focused on the genetic background of the Uygur group, however, the patrilineal descent of the group is still unclear. In this study, we investigated the genetic diversity of 24 Y-STR loci in the Uygur group and analyzed the population differentiations as well as the genetic relationships between the Uygur group and other previously reported populations using 17 Y-filer loci. According to haplotypic analysis of the 24 Y-STR loci in 109 Uygur individuals, 104 different haplotypes were obtained, 99 of which were unique. The haplotypic diversity and discrimination capacity of these 24 Y-STR loci in Uygur group were 0.9992 and 0.9541, respectively. An additional 7 loci (DYS388, DYS444, DYS447, DYS449, DYS522, and DYS527a,b) showed high genetic diversity and improved the overall discrimination capacity of the 24 Y-STR system. Pairwise Fst and neighbor-joining analysis showed that the Uygur group was genetically close to the Han populations from different regions.

Developmental validation of the AGCU 21+1 STR kit: a novel multiplex assay for forensic application.

In this study, we describe the developmental validation assay performed on a novel designed STR multiplex system, AGCU 21+1 STR kit. This kit contains a sex-determining locus amelogenin and 21 noncombined DNA index system STR loci, that are, D6S474, D12ATA63, D22S1045, D10S1248, D1S1677, D11S4463, D1S1627, D3S4529, D2S441, D6S1017, D4S2408, D19S433, D17S1301, D1GATA113, D18S853, D20S482, D14S1434, D9S1122, D2S1776, D10S1435, and D5S2500. The 21+1 kit was validated by a series of tests including optimized PCR conditions, sensitivity, precision and accuracy, stutter ratio, DNA mixture, inhibitors, and species specificity according to the revised validation guidelines issued by the Scientific Working Group on DNA Analysis Methods (SWGDAM). Our results in this study show that the kit is a useful tool for forensic application.

Single nucleotide polymorphisms of mitochondrial DNA HVS-I and HVS-II in Chinese Bai ethnic group.

For forensic and population genetic purposes, a total of 125 unrelated volunteers' blood samples were collected from Chinese Bai ethnic minority group to analyze sequence variation of two hypervariable segments (HVS-I and HVS-II) in the mitochondrial DNA control region. Comparing the HVS-I and HVS-II sequences of the 125 Chinese Bais to the Anderson reference sequence, we found 86 polymorphic loci in HVS-I and 40 in HVS-II in mitochondrial DNA sequences of the Chinese Bai ethnic minority group, which defined 93 and 53 different haplotypes, respectively. Haplotype diversity and the mean pairwise differences were 0.992 ± 0.003 and 6.553 in HVS-I, and 0.877 ± 0.027 and 2.407 in HVS-II, respectively. We defined four macrohaplogroups R, M, N and D with the proportions ranging from 9.6% to 40.0%. With the analysis of the hypervariable domain from nucleotide 16 180-16 193 in HVS-I, our study revealed new haplotypes of sequence variations. In addition, the Fst metric, phylogenetic tree, and principal component analysis demonstrated a close genetic relationship between the Bai group and Chinese Han populations from South China, Changsha, and Guangdong. The results support that the Bai group is a multiorigin ethnic minority that has merged with the Chinese Han population.

Forensic evaluation and population genetic study of 30 insertion/deletion polymorphisms in a Chinese Yi group.

Insertion/deletion polymorphisms have become a research hot spot in forensic science due to their tremendous potential in recent years. In the present study, we investigated 30 indel loci in a Chinese Yi ethnic group. The allele frequencies of the short allele of the 30 indel loci were in the range of 0.1025-0.9221. The power of discrimination values were observed ranging from to 0.2630 (HLD111 locus) to 0.6607 (HLD70 locus) and probability of exclusion values ranged from 0.0189 (HLD111 locus) to 0.2343 (HLD56 locus). The combined power of discrimination and power of exclusion for 30 loci in the studied Yi group were 0.99999999995713 and 0.97746, respectively, which showed tremendous potential for forensic personal identification in the Yi group. Moreover, the DA distances, phylogenetic tree, principal component analysis, and cluster analysis showed the Yi group had close genetic relationships with the Tibetan, South Korean, Chinese Han, and She groups.

Structural polymorphism analysis of Chinese Mongolian ethnic group revealed by a new STR panel: genetic relationship to other groups.

Mongolian is the eighth largest ethnic minority on Chinese population data according to the 2010 census. In the present study, we presented the first report about the allelic frequencies and forensic statistical parameters at the 21 new STRs and analyzed linkage disequilibrium of pairwise loci in the Mongolian ethnic minority, China. Hardy-Weinberg equilibrium tests demonstrated no significant deviations except for the D1S1627 locus. The cumulative power of discrimination and power of exclusion of all the loci are 0.9999999999999999992576 and 0.9999997528, respectively. The results of analysis of molecular variance showed that significant differences between the Mongolian and the other eight populations were found at 1-9 STR loci. In population genetics, the results of principal component analysis, structure analysis, and phylogenetic reconstruction analysis indicated shorter genetic distance between the Mongolian group and the Ningxia Han. All the results suggest that the 21 new STR loci will contribute to Chinese population genetics and forensic caseworks in the Mongolian group.

Genetic profile characterization and population study of 21 autosomal STR in Chinese Kazak ethnic minority group.

Short tandem repeat loci have been recognized as useful tools in the routine forensic application and in recent decades, more and more new short tandem repeat (STR) loci have been constantly discovered, studied, and applied in forensic caseworks. In this study, we investigated the genetic polymorphisms of 21 STR loci in the Kazak ethnic minority as well as the genetic relationships between the Kazak ethnic minority and other populations. Allelic frequencies of 21 STR loci were obtained from 114 unrelated healthy Kazak individuals in the Ili Kazak Autonomous Prefecture, Xinjiang Uigur Autonomous Region of China. We observed a total of 159 alleles in the group with the allelic diversity values ranging from 0.0044 to 0.5088. The highest polymorphism was found at D19S433 locus and the lowest was found at D1S1627. Statistical analysis of the generated data indicated no deviation from Hardy-Weinberg equilibriums at all 21 STR loci. In order to estimate the population differentiation, allelic frequencies of all STR loci of the Kazak were compared with those of other neighboring populations using analysis of molecular variance method. Statistically significant differences were found between the studied population and other populations at 2-7 STR loci. A neighbor-joining tree was constructed based on allelic frequencies of the 21 STR loci and phylogenetic analysis indicates that the Kazak has a close genetic relationship with the Uigur ethnic group. The present results may provide useful information for forensic sciences and population genetics studies, and can also increase our understanding of the genetic background of this group. The present findings showed that all the 21 STR loci are highly genetically polymorphic in the Kazak group, which provided valuable population genetic data for the genetic information study, forensic human individual identification, and paternity tests.

Genetic diversity and haplotype structure of 24 Y-chromosomal STR in Chinese Hui ethnic group and its genetic relationships with other populations.

In the present study, 24 Y-chromosomal short tandem repeat (Y-STR) loci were analyzed in 115 unrelated Hui male individuals from Haiyuan county or Tongxin county, Ningxia Hui Autonomous Region, China, to evaluate the forensic application of the 24 STR loci and to analyze interpopulation differentiations by making comparisons between the Hui group data and previously published data of other 13 populations. A total of 115 different haplotypes were observed on these 24 Y-STR loci. The gene diversities ranged from 0.4049 (DYS437) to 0.9729 (DYS385a, b). The overall haplotype diversity was 1 at AGCU 24 Y-STR loci level, while the values were reduced to 0.999237, 0.996949, and 0.996644 at the Y-filer 17 loci, 11 Y-STR loci of extended haplotype and 9 Y-STR loci of minimal haplotype levels, respectively; whereas, haplotype diversity for additional 7 loci (not included in Y-filer 17 loci) was 0.995271. The pairwise FST , multidimensional scaling plot and neighbor-joining tree indicated the Hui group had the closest genetic relationship with Sala in the paternal lineage in the present study. In summary, the results in our study indicated the 24 Y-STRs had a high level of polymorphism in Hui group and hence could be a powerful tool for forensic application and population genetic study.

Diversity study of 12 X-chromosomal STR loci in Hui ethnic from China.

X-chromosomal STRs (X-STRs) have been used as complements of autosomal STR application in recent years. In this work, we present population genetic data of 12 X-STRs including DXS101, DXS10159, DXS10162, DXS10164, DXS6789, DXS7133, DXS7423, DXS7424, DXS8378, DXS981, GATA165B12, and GATA31E08 loci in a sample of 231 unrelated healthy individuals from the Hui ethnic group in Ningxia Hui Autonomous Region, China. Allelic frequencies of the 12 X-STR loci and haplotypic frequencies of the reported linkage groups (DXS7424-DXS101 and DXS10159-DXS10164-DXS10162) were investigated in the group, respectively. No STR loci showed significant deviations from the Hardy-Weinberg equilibriums and no linkage disequilibriums of pairwise loci were found after Bonferroni correction, respectively. A combined power of discrimination in female individuals was 0.999999999985 and that in male individuals was 0.99999967, respectively. The combined mean exclusion chance in deficiency cases, normal trios and duo cases were 0.999934, 0.995754, and 0.999796, respectively. Significant differences were observed from 0 to 8 loci, when making comparisons between the data of Hui ethnic group and previously reported data from other 16 populations. The results indicated the new panel of 12 X-STR loci might be useful for forensic science application.

Polymorphic distribution and forensic effectiveness study of eight miniSTR in Chinese Uyghur ethnic group.

We obtained the allelic frequencies and forensic efficiency data for eight mini short tandem repeat loci including Penta E, D12S391, D6S1043, D2S1338, D19S433, CSF1PO, Penta D and D19S253 loci from a sample of 128 unrelated Uyghur individuals from China. The amplification products of the eight STR loci are <240 bp in size. A total of 94 alleles were observed and the corresponding allelic frequencies ranged from 0.0039 to 0.3438 in the present study. Observed genotype distributions for each locus do not show deviations from Hardy-Weinberg equilibrium expectations. The combined power of discrimination, combined power of exclusion and combined matching probability of the eight STR loci equaled to 0.999999999963373, 0.9997770 and 3.6627 × 10(-11), respectively. Because of the small fragment length of PCR products and the high degree of polymorphisms, the eight STR loci are highly beneficial for the forensic analysis of degraded DNA samples which are commonly observed in forensic cases. The STR data of the Uyghur group were compared with the previously published population STR data of other groups from different ethnic or areas, and significant differences were observed among these groups at some loci.

Population genetics and forensic efficiency of twenty-one novel microsatellite loci of Chinese Yi ethnic group.

In this study, we investigated polymorphic distributions of allelic frequencies and forensic genetic parameters of 21 novel autosomal microsatellite loci from 110 unrelated healthy individuals of Chinese Yi ethnic group. Expected heterozygosity, power of discrimination, and polymorphic information content ranged from 0.617 to 0.812, 0.777 to 0.936 and 0.560 to 0.790. The microsatellite loci showed high forensic efficiency. The total discrimination power and cumulate probability of exclusion were 0.99999999999999999986902 and 0.999998818, respectively. Locus-by-locus allelic frequencies were compared using analysis of molecular variance (AMOVA) method, and the statistically significant differences were observed between Yi group and Russian, Tujia, Kazak, Bai, Ningxia Han, Salar, Tibetan, and Uigur groups at 5, 6, 7, 7, 7, 8, 12, and 13 loci, respectively. The results of genetic distance comparisons, genetic structure analyses, and principal component analysis all indicated that the Yi group showed relatively short genetic relationships with Russian, Salar, and Bai group. The experimental results showed that the 21 loci in the multiplex system provided highly polymorphic information and forensic efficiency for forensic individual identification and paternity testing, also basic population data for population genetics and anthropological research.

The impact of gender, age and tissues in vitro on estimating postmortem interval by FTIR spectroscopy.

To investigate the influence of such individual factors as gender, age and tissues in vitro to the postmortem interval (PMI) by the Fourier transform infrared (FTIR) spectrometer in animal experiments. SD rats were classified into male and female groups, different age groups (21-day, 42-day and 63-day group), and tissues in vitro and in vivo groups. The rats were sacrificed by cervical dislocation, whose bodies were kept in a controlled environmental chamber set at (20+/-2) degrees C and 50% humidity. The liver, kidney, spleen, myocardium, brain, lung and skeletal muscle tissues were collected for measurement from time zero to 48 h postmortem. With the change of PMI, no obvious changes were found in the main FTIR absorbance peaks and their ratios at different time points. All the experimental groups showed no significant changes when compared with the controls. The gender, age and tissues in vitro were not found to be contributing factors in the estimation of PMI via FTIR spectroscopy.

Automatic sex estimation using deep convolutional neural network based on orthopantomogram images.

Sex estimation is very important in forensic applications as part of individual identification. Morphological sex estimation methods predominantly focus on anatomical measurements. Based on the close relationship between sex chromosome genes and facial characterization, craniofacial hard tissues morphology shows sex dimorphism. In order to establish a more labor-saving, rapid, and accurate reference for sex estimation, the study investigated a deep learning network-based artificial intelligence (AI) model using orthopantomograms (OPG) to estimate sex in northern Chinese subjects. In total, 10703 OPG images were divided into training (80%), validation (10%), and test sets (10%). At the same time, different age thresholds were selected to compare the accuracy differences between adults and minors. The accuracy of sex estimation using CNN (convolutional neural network) model was higher for adults (90.97%) compared with minors (82.64%). This work demonstrated that the proposed model trained with a large dataset could be used in automatic morphological sex-related identification with favorable performance and practical significance in forensic science for adults in northern China, while also providing a reference for minors to some extent.