Machine Learning Supported the Modified Gustafson's Criteria for Dental Age Estimation in Southwest China.

Adult age estimation is one of the most challenging problems in forensic science and physical anthropology. In this study, we aimed to develop and evaluate machine learning (ML) methods based on the modified Gustafson's criteria for dental age estimation. In this retrospective study, a total of 851 orthopantomograms were collected from patients aged 15 to 40 years old. The secondary dentin formation (SE), periodontal recession (PE), and attrition (AT) of four mandibular premolars were analyzed according to the modified Gustafson's criteria. Ten ML models were generated and compared for age estimation. The partial least squares regressor outperformed other models in males with a mean absolute error (MAE) of 4.151 years. The support vector regressor (MAE = 3.806 years) showed good performance in females. The accuracy of ML models is better than the single-tooth model provided in the previous studies (MAE = 4.747 years in males and MAE = 4.957 years in females). The Shapley additive explanations method was used to reveal the importance of the 12 features in ML models and found that AT and PE are the most influential in age estimation. The findings suggest that the modified Gustafson method can be effectively employed for adult age estimation in the southwest Chinese population. Furthermore, this study highlights the potential of machine learning models to assist experts in achieving accurate and interpretable age estimation.

Adult age estimation from the sternum using maximum intensity projection images of CT and data mining in a Chinese population.

This study aimed to explore and develop data mining models for adult age estimation based on CT reconstruction images from the sternum. Maximum intensity projection (MIP) images of chest CT were retrospectively collected from a modern Chinese population, and data from 2700 patients (1349 males and 1351 females) aged 20 to 70 years were obtained. A staging technique within four indicators was applied. Several data mining models were established, and mean absolute error (MAE) was the primary comparison parameter. The intraobserver and interobserver agreement levels were good. Within internal validation, the optimal data mining model obtained the lowest MAE of 9.08 in males and 10.41 in females. For the external validation (N = 200), MAEs were 7.09 in males and 7.15 in females. In conclusion, the accuracy of our model for adult age estimation was among similar studies. MIP images of the sternum could be a potential age indicator. However, it should be combined with other indicators since the accuracy level is still unsatisfactory.

Machine learning and deep learning enabled age estimation on medial clavicle CT images.

The medial clavicle epiphysis is a crucial indicator for bone age estimation (BAE) after hand maturation. This study aimed to develop machine learning (ML) and deep learning (DL) models for BAE based on medial clavicle CT images and evaluate the performance on normal and variant clavicles. This study retrospectively collected 1049 patients (mean± SD: 22.50±4.34 years) and split them into normal training and test sets, and variant training and test sets. An additional 53 variant clavicles were incorporated into the variant test set. The development stages of normal MCE were used to build a linear model and support vector machine (SVM) for BAE. The CT slices of MCE were automatically segmented and used to train DL models for automated BAE. Comparisons were performed by linear versus ML versus DL, and normal versus variant clavicles. Mean absolute error (MAE) and classification accuracy was the primary parameter of comparison. For BAE, the SVM had the best MAE of 1.73 years, followed by the commonly-used CNNs (1.77-1.93 years), the linear model (1.94 years), and the hybrid neural network CoAt Net (2.01 years). In DL models, SE Net 18 was the best-performing DL model with similar results to SVM in the normal test set and achieved an MAE of 2.08 years in the external variant test. For age classification, all the models exhibit superior performance in the classification of 18-, 20-, 21-, and 22-year thresholds with limited value in the 16-year threshold. Both ML and DL models produce desirable performance in BAE based on medial clavicle CT.

Automated bone age assessment from knee joint by integrating deep learning and MRI-based radiomics.

Bone age assessment (BAA) is a crucial task in clinical, forensic, and athletic fields. Since traditional age estimation methods are suffered from potential radiation damage, this study aimed to develop and evaluate a deep learning radiomics method based on multiparametric knee MRI for noninvasive and automatic BAA. This retrospective study enrolled 598 patients (age range,10.00-29.99 years) who underwent MR examinations of the knee joint (T1/T2*/PD-weighted imaging). Three-dimensional convolutional neural networks (3D CNNs) were trained to extract and fuse multimodal and multiscale MRI radiomic features for age estimation and compared to traditional machine learning models based on hand-crafted features. The age estimation error was greater in individuals aged 25-30 years; thus, this method may not be suitable for individuals over 25 years old. In the test set aged 10-25 years (n = 95), the 3D CNN (a fusion of T1WI, T2*WI, and PDWI) demonstrated the lowest mean absolute error of 1.32 ± 1.01 years, which is higher than that of other MRI modalities and the hand-crafted models. In the classification for 12-, 14-, 16-, and 18- year thresholds, accuracies and the areas under the ROC curves were all over 0.91 and 0.96, which is similar to the manual methods. Visualization of important features showed that 3D CNN estimated age by focusing on the epiphyseal plates. The deep learning radiomics method enables non-invasive and automated BAA from multimodal knee MR images. The use of 3D CNN and MRI-based radiomics has the potential to assist radiologists or medicolegists in age estimation.

Forensic age estimation in adults by pubic bone mineral density using multidetector computed tomography.

Radiology plays a crucial role in forensic anthropology for age estimation. However, most studies rely on morphological methods. This study aims to investigate the feasibility of using pubic bone mineral density (BMD) as a new age estimation method in the Chinese population. 468 pubic bone CT scans from living individuals in a Chinese hospital aged 18 to 87 years old were used to measure pubic BMD. The BMD of the bilateral pubic bone was measured using the Mimics software on cross-sectional CT images and the mean BMD of the bilateral pubic bone was also calculated. Regression analysis was performed to assess the correlation between pubic BMD and chronological age and to develop mathematical models for age estimation. We evaluated the accuracy of the best regression model using an independent validation sample by calculating the mean absolute error (MAE). Among all established models, the cubic regression model had the highest R2 value in both genders, with R2 = 0.550 for males and R2 = 0.634 for females. The results of the best model test showed that the MAE for predicting age using pubic BMD was 8.66 years in males and 7.69 years in females. This study highlights the potential of pubic BMD as a useful objective indicator for adult age estimation and could be used as an alternative in forensic practice when other better indicators are lacking.

Deep learning enables automatic adult age estimation based on CT reconstruction images of the costal cartilage.

OBJECTIVE: Adult age estimation (AAE) is a challenging task. Deep learning (DL) could be a supportive tool. This study aimed to develop DL models for AAE based on CT images and compare their performance to the manual visual scoring method. METHODS: Chest CT were reconstructed using volume rendering (VR) and maximum intensity projection (MIP) separately. Retrospective data of 2500 patients aged 20.00-69.99 years were obtained. The cohort was split into training (80%) and validation (20%) sets. Additional independent data from 200 patients were used as the test set and external validation set. Different modality DL models were developed accordingly. Comparisons were hierarchically performed by VR versus MIP, single-modality versus multi-modality, and DL versus manual method. Mean absolute error (MAE) was the primary parameter of comparison. RESULTS: A total of 2700 patients (mean age = 45.24 years ± 14.03 [SD]) were evaluated. Of single-modality models, MAEs yielded by VR were lower than MIP. Multi-modality models generally yielded lower MAEs than the optimal single-modality model. The best-performing multi-modality model obtained the lowest MAEs of 3.78 in males and 3.40 in females. On the test set, DL achieved MAEs of 3.78 in males and 3.92 in females, which were far better than the MAEs of 8.90 and 6.42 respectively, for the manual method. For the external validation, MAEs were 6.05 in males and 6.68 in females for DL, and 6.93 and 8.28 for the manual method. CONCLUSIONS: DL demonstrated better performance than the manual method in AAE based on CT reconstruction of the costal cartilage. CLINICAL RELEVANCE STATEMENT: Aging leads to diseases, functional performance deterioration, and both physical and physiological damage over time. Accurate AAE may aid in diagnosing the personalization of aging processes. KEY POINTS: • VR-based DL models outperformed MIP-based models with lower MAEs and higher R2 values. • All multi-modality DL models showed better performance than single-modality models in adult age estimation. • DL models achieved a better performance than expert assessments.

Major Adverse Cardiovascular Events Among Obese Patients with Diabetes After Metabolic and Bariatric Surgery: a Meta-analysis of Matched Cohort and Prospective Controlled Studies with 122,361 Participates.

PURPOSE: Metabolic and bariatric surgery (MBS) can exert effective function on glycemic control. The present study aimed to estimate the risk of MACE among obese patients with diabetes after MBS. MATERIALS AND METHODS: Systematic search of PubMed, Embase, Medline, and Web of Science was performed for studies published before 20th February 2023. The odds ratio (OR) corresponding to the 95% confidence interval (95% CI) was used to assess the outcome. The statistical heterogeneity among studies was assessed with the Q-test and I2 statistics. RESULTS: Fifteen cohort studies with 122,361 obese patients with diabetes were available for analysis. Our meta-analysis found significantly decreased morbidity and mortality of MACE (OR = 0.65, 95% CI = 0.59-0.72, I2 = 62.8% for morbidity, OR = 0.49, 95% CI = 0.36-0.67, I2 = 68.7% for mortality). Subgroup analysis revealed MBS decreased cerebrovascular disease, coronary artery disease, atrial fibrillation, heart failure, myocardial infarction, and stroke risk. CONCLUSION: Our meta-analysis indicated that MBS for obese patients with diabetes is beneficial to decreasing MACE risk. Moreover, further studies estimating the functional effect may eventually provide a better and comprehensive understanding of the effect on different populations.

Forensic age prediction and age classification for critical age thresholds via 3.0T magnetic resonance imaging of the knee in the Chinese Han population.

OBJECTIVES: To assess the performance of knee MRI for forensic age prediction and classification for 12-, 14-, 16-, and 18-year thresholds. METHODS: The ossification stages of distal femoral epiphyses and proximal tibial epiphyses were assessed using an integrated staging system by Schmeling et al. and Kellinghaus et al. for knee 3.0T MRI with T1-weighted turbo spin-echo (T1-TSE) in sagittal orientation among 852 Chinese Han individuals (483 males and 369 females) aged 7-30 years. Regression models for age prediction were constructed and their performances were evaluated based on mean absolute deviation (MAD) values. In addition, the performances of age classification were assessed using receiver operating characteristic (ROC) analyses. RESULTS: The intra- and inter-observer agreement levels were very good (κ > 0.80). The complete fusion of those two types of epiphyses took place before 18.0 years in our study participants. The minimum MAD values were 2.51 years (distal femur) and 2.69 years (proximal tibia) in males, and 2.75 years (distal femur) and 2.87 years (proximal tibia) in females. The specificity values of constructed prediction models were all above 90% for the 12-, 14-, and 16-year thresholds, compared to the 74.8-84.6% for the 18-year threshold. Better performances of age prediction and classification were observed in males by distal femoral epiphyses. CONCLUSIONS: Ossification stages via 3.0T MRI of the knee with T1-TSE sequence using an integrated staging system could be a reliable noninvasive method for age prediction or for age classification for 12-, 14-, and 16-year thresholds, especially in males by distal femoral epiphyses. However, assessments based on the full bony fusion of the distal femoral epiphysis and proximal tibial epiphysis seemed not reliable for age classification for the 18-year threshold in the Chinese Han population.

ERCC5 quantification for age estimation of peripheral blood in a Chinese Han population.

Age estimation is an important topic of human identification in forensic practice, especially coming to biological samples in crime scene, such as blood, saliva, semen. As rate-limiting enzyme in Nucleotide excision repair (NER) that was associated with aging, Excision repair cross-complementation group 5 (ERCC5) was considered to be a candidate biomarker for individual age estimation. The ERCC5 mRNA and protein expression levels association with age have been demonstrated in our previous study. However, very little is known about relationship DNA-based quantification of ERCC5 with age. In this study, we detected ERCC5 level in peripheral blood from a Chinese Han population by SYBR qPCR assay to gain better insight into the quantitative relationship with age. The results showed ERCC5 level declined with individual age with a negative correlation(r = -0.8, R2 = 0.63, P < 0.001). The data model for age estimation based on ERCC5 level was Y = -31.352X + 14.436 ± 10.28 (Y: age, year; X: CqTBP-CqERCC5; standard error: year). The accuracy about the data model for age estimation was about 73.33%. The mean absolute difference (MAD) values were 8.22, 8.09 and 8.38 in total, male and female, respectively. Furthermore, ERCC5 quantification for age estimation was also applicable for stored blood samples under low temperature up to 6 months. It was suggested that the ERCC5 quantification was expected to be a valuable additional method for individual age estimation, especially in cases where traditional morphologic method is absent or inefficient in forensic practice.

Estimation of stature from radiographically determined lower limb bone length in modern Chinese.

To develop population - specific stature prediction equations from measurements of the lower limb bone in a contemporary Chinese. 303 individuals of Han group in Western China, including 201 females and 102 males were collected. The study sample was randomly divided into two subgroups. A calibration sample, which consisted of 171 females and 87 males, was used to develop the regression formula. A validation sample comprising the remaining 30 female and 15 male individuals was then used to test the predictive accuracy of the established formula. The regression equations were developed from intact bones and fragments of the femur, tibia and fibula, the maximum lengths of femur, tibia, and fibula were highly correlated with the stature. The maximum length of femur provide the most accurate result with the prediction accuracy of 3.84 cm for unknown sex, 4.00 cm in the male group, 3.45 cm in the female group, 3.61 cm in the group with age no more than 45, 3.45 cm in the group with age above 45. Moreover, the multiple regression equations were developed, and they portray a more accurate stature in instances in which the femur, tibia and fibula are available. This paper provides indications that the femur, tibia and fibula are important bones for stature estimation and they could be effectively used in forensic cases.

Estimation of sex based on patella measurements in a contemporary Chinese population using multidetector computed tomography: An automatic measurement method.

Sex estimation is an important part of creating a biological profile, and ultimately assisting in creating a presumptive identification of unidentified skeletal remains. However, manual methods of anthropometric are time-consuming and prone to observer variability. The present study is an attempt to estimation of sex from automatic measurement of patella by multidetector computed tomography (MDCT) in a contemporary Chinese population. Four measurements for every patella, including maximum height (MAXH), maximum breadth (MAXB), maximum thickness (MAXT) and patellar volume (PV), were automatically provided by the software from CT image of 300 Chinese. The sample is composed of 156 males and 144 females with an average age of 41.44 and 45.68 years, respectively. The statistical analyses showed that all variables were sexually dimorphic. Receiver operating characteristic (ROC) analysis was performed to estimate sex from patella. The univariate analysis of each patellar parameter yielded a sex classification accuracy rate of 73.1% to 85.7%. The classification accuracy rates of sex estimation using the combination of the patellar parameters are 81.9% to 91.6%. This paper provides indications that the patella is important bone for sex estimation and it may be used as an alternative in forensic cases when the skull and pelvis are unavailable.

Age estimation based on magnetic resonance imaging of the ankle joint in a modern Chinese Han population.

OBJECTIVES: To expand the database on magnetic resonance imaging (MRI) analysis of distal tibial and calcaneal epiphyses as proposed by Saint-Martin et al. and investigate a more elaborate staging technique to establish regression models for age estimation in a modern Chinese Han population. MATERIALS AND METHODS: T1-weighted ankle MRIs were retrospectively collected from April 2008 to July 2019, and data from 590 individuals (372 males and 218 females; aged from 8 to 25 years old) were obtained. One-sided sagittal images were assessed because data from both sides were considered coincidental, as no significant differences were found (P > 0.05). Three-stage and six-stage staging techniques were applied separately and subsequently compared. A subset was re-assessed a second time and by a different observer. Regression models were established accordingly. RESULTS: Our results showed very good repeatability and consistency of two staging techniques (all Cohen's kappa values were more than 0.8). By comparison, the values of the coefficient of determination (R2) of the six-stage technique were generally higher than those of the three-stage technique. Compared with the distal tibia and two ankle bones combined, the calcaneus decreased the mean absolute deviation (MAD) with the six-stage technique. In males, incorporating only the calcaneus resulted in a MAD of 2.15 years, with correct classification rates of 87.5% adults and 50.0% among minors. In females, the corresponding results were 1.67 years, 100.0%, and 44.4%, respectively. CONCLUSIONS: The six-stage technique may outperform the three-stage technique in MRI analysis of ankle bones for age estimation, while age estimation based on the calcaneus may perform better than that based on the distal tibia or both ankle bones in a modern Chinese Han population.

Estimation of stature and sex from skull measurements by multidetector computed tomography in Chinese.

The present study is an attempt to estimation of stature and sex from skull measurements by multidetector computed tomography (MDCT) in contemporary Chinese. In the present study, fifteen measurements for every skull were taken from CT image of 382 Chinese. The sample was composed of 200 males and 182 females with an average age of 47 and 46 years, respectively. Discriminant function was used in sex determination and regression analysis was used in stature estimation from skull measurements. The stepwise analysis of all measurements yielded a sex classification accuracy rate of 89.3%. The classification accuracy rates of the univariate discriminant function analyses were from 50.5% to 84.8%. For stature estimation, the standard error of estimate (SEE) ranged from 5.072 to 6.355 cm for male, from 5.090 to 5.829 cm for female, respectively. This study is the first to provide a metric and statistical characterization of the skull in contemporary Chinese, and indicates that it is feasible to sex estimation by skull measurement. Furthermore, the equations presented for stature estimation in this study may be used as alternatives in forensic cases, particularly in cases where better predictors such as the long bones are not available.

Stature estimation from the combination of clavicle, scapula, and sternum from 3D-VRT images in Chinese population.

The objective of the present study was to generate multiple regression models for stature estimation on the basis the combination of the clavicle, scapula and sternum derived from 3D-VRT images in Chinese population. The study sample comprised 363 individuals from China, including 159 females and 204 males, with documented ages between 19 and 82 years. Separate multiple linear regression equations for estimating stature on the basis of the measurements from the clavicle, scapula and sternum were then devised for males and females. For assessing the correlation between the stature and measurements of the clavicle, scapula and sternum, the Pearson's correlation coefficient was calculated and its significance was tested by Students t test. Finally, the multiple regression equations calculated from the measurements of the clavicle, scapula and sternum in relation to stature for each sex were established in the present study. The accuracy of stature prediction ranged from 4.777 to 5.313 cm for male and from 4.388 to 4.658 cm for female. In conclusion, the present results provide indications that the combination of the clavicle, scapula and sternum should be used as alternatives for stature estimation, and the multiple equations presented for stature estimation seem to be a more accurate than the equations from single bone.

Estimation of stature and sex from sacrum and coccyx measurements by multidetector computed tomography in Chinese.

The present study is an attempt to estimation of stature and sex from sacrum and coccyx measurements by multidetector computed tomography (MDCT) in a contemporary Chinese population. Nine measurements for every sacrum and coccyx were taken from CT image of 350 Chinese. The sample is composed of 190 males and 160 females with an average age of 55 and 50 years, respectively. Discriminant function was used in sex estimation and regression analysis was used in stature estimation from these two bones. The stepwise analysis of all measurements yielded a sex classification accuracy rate of 84.9%. The classification accuracy rates of the univariate discriminant function analyses are 58.3%-76.9%. For stature estimation, the accuracy of stature prediction ranged from 4.891 to 6.107 cm for male, from 4.474 to 5.606 cm for female, respectively. This paper provides indications that the sacrum and coccyx are important bones for sex estimation and they could be effectively used as alternatives in forensic cases when the skull and pelvis are unavailable. Furthermore, the regression equations presented in this study may be useful for forensic estimation of the stature of Chinese individuals, particularly in cases where better predictors such as the long bones are not available.