The BoneXpert method for automated determination of skeletal maturity

Comparison of Bone Agesin Early Puberty: Computerized Greulich-Pyle Based Bone Age vs. Sauvegrain Method

Purpose

To compare the computerized Greulich-Pyle based bone age with elbow bone age.

Materials and Methods

A total of 2126 patients (1525 girls; 601 boys) whose elbow bone age was within the evaluable range by the Sauvegrain method, and who simultaneously underwent hand radiography, were enrolled in the study. The 1st-bone age and VUNO score of the hand were evaluated using VUNOMed-BoneAge software. The correlation between the hand and elbow bone age was analyzed according to the child's gender and the probability of 1st-bone age.

Results

The correlation between VUNO score and elbow bone age (r = 0.898) was higher than the correlation between 1st-bone age and elbow bone age (r = 0.879). Moreover, the VUNO score showed a better correlation with the elbow bone age in patients with a 1st-bone age probability of less than 70%, or in girls. Elbow bone age was more advanced compared to hand bone age, and this difference increased until the middle of puberty and gradually decreased in the latter half.

Conclusion

The computerized Greulich-Pyle based hand bone age showed a significant correlation with the elbow bone age at puberty. However, since the elbow bone age tends to advance faster than the hand bone age, caution is required while judging the bone age during puberty.

Skeletal age assessed by TW2 using 20-bone, carpal and RUS score systems: Intra-observer and inter-observer agreement among male pubertal soccer players

The purpose of this study was to determine intra- and inter-observer agreement for the three skeletal ages derived from the TW2 method among male pubertal soccer players. The sample included 142 participants aged 11.0–15.3 years. Films of the left hand-wrist were evaluated twice by each of two observers. Twenty bones were rated and three scoring systems used to determine SA adopting the TW2 version: 20-bone, CARPAL and RUS. Overall agreement rates were 95.1% and 93.8% for, respectively, Observer A and Observer B. Although, agreement rates between observers differed for 13 bones (5 carpals, metacarpal-I, metacarpal-III, metacarpal-V, proximal phalanges-I, III and V, distal phalanx-III), intra-class correlationa were as follows: 0.990 (20-bone), 0.969 (CARPAL), and 0.988 (RUS). For the three SA protocols, BIAS was negligible: 0.02 years (20-bone), 0.04 years (CARPAL), and 0.03 years (RUS). Observer-associated error was not significant for 20-bone SA (TEM = 0.25 years, %CV = 1.86) neither RUS SA (TEM = 0.31 years, %CV = 2.22). Although the mean difference for CARPAL SAs between observers (observer A: 12.48±1.18 years; observer B: 12.29±1.24 years; t = 4.662, p<0.01), the inter-observer disagreement had little impact (TEM: 0.34 years: %CV: 2.78). The concordance between bone-specific developmental stages seemed was somewhat more problematic for the carpals than for the long bones. Finally, when error due to the observer is not greater than one stage and the replicated assignments had equal probability for being lower or higher compared to initial assignments, the effect on SAs was trivial or small.

A Cascade Model with Prior Knowledge for Bone Age Assessment

Bone age is commonly used to reflect growth and development trends in children, predict adult heights, and diagnose endocrine disorders. Nevertheless, the existing automated bone age assessment (BAA) models do not consider the nonlinearity and continuity of hand bone development simultaneously. In addition, most existing BAA models are based on datasets from European and American children and may not be applicable to the developmental characteristics of Chinese children. Thus, this work proposes a cascade model that fuses prior knowledge. Specifically, a novel bone age representation is defined, which incorporates nonlinear and continuous features of skeletal development and is implemented by a cascade model. Moreover, corresponding regions of interest (RoIs) based on RUS-CHN were extracted by YOLO v5 as prior knowledge inputs to the model. In addition, based on MobileNet v2, an improved feature extractor was proposed by introducing the Convolutional Block Attention Module and increasing the receptive field to improve the accuracy of the evaluation. The experimental results show that the mean absolute error (MAE) is 4.44 months and significant correlations with the reference bone age is (r = 0.994, p < 0.01); accuracy is 94.04% for ground truth within ±1 year. Overall, the model design adequately considers hand bone development features and has high accuracy and consistency, and it also has some applicability on public datasets, showing potential for practical and clinical applications.

A Global-Local Feature Fusion Convolutional Neural Network for Bone Age Assessment of Hand X-ray Images

Bone age assessment plays a critical role in the investigation of endocrine, genetic, and growth disorders in children. This process is usually conducted manually, with some drawbacks, such as reliance on the pediatrician’s experience and extensive labor, as well as high variations among methods. Most deep learning models use one neural network to extract the global information from the whole input image, ignoring the local details that doctors care about. In this paper, we propose a global-local feature fusion convolutional neural network, including a global pathway to capture the global contextual information and a local pathway to extract the fine-grained information from local patches. The fine-grained information is integrated into the global context information layer-by-layer to assist in predicting bone age. We evaluated the proposed method on a dataset with 11,209 X-ray images with an age range of 4–18 years. Compared with other state-of-the-art methods, the proposed global-local network reduces the mean absolute error of the estimated ages to 0.427 years for males and 0.455 years for females; the average accuracy rate is within 6 months and 12 months, reaching 70% and 91%, respectively. In addition, the effectiveness and rationality of the model were verified on a public dataset.

Near-Adult Heights and Adult Height Predictions Using Automated and Conventional Greulich-Pyle Bone Age Determinations in Children with Chronic Endocrine Diseases

OBJECTIVES

To validate adult height predictions (BX) using automated and Greulich-Pyle bone age determinations in children with chronic endocrine diseases.

METHODS

Heights and near-adult heights were measured in 82 patients (48 females) with chronic endocrinopathies at the age of 10.45 ± 2.12 y and at time of transition to adult care (17.98 ± 3.02 y). Further, bone age (BA) was assessed using the conventional Greulich-Pyle (GP) method by three experts, and by BoneXpert™. PAH were calculated using conventional BP tables and BoneXpert™.

RESULTS

The conventional and the automated BA determinations revealed a mean difference of 0.25 ± 0.72 y (p = 0.0027). The automated PAH by BoneXpert™ were 156.26 ± 0.86 cm (SDS - 2.01 ± 1.07) in females and 171.75 ± 1.6 cm (SDS - 1.29 ± 1.06) in males, compared to 153.95 ± 1.12 cm (SDS - 2.56 ± 1.5) in females and 169.31 ± 1.6 cm (SDS - 1.66 ± 1.56) in males by conventional BP, respectively and in comparison to near-adult heights 156.38 ± 5.84 cm (SDS - 1.91 ± 1.15) in females and 168.94 ± 8.18 cm (SDS - 1.72 ± 1.22) in males, respectively.

CONCLUSION

BA ratings and adult height predictions by BoneXpert™ in children with chronic endocrinopathies abolish rater-dependent variability and enhance reproducibility of estimates thereby refining care in growth disorders. Conventional methods may outperform automated analyses in specific cases.

Applicability of the Greulich and Pyle bone age standards to Taiwanese children: A Taipei experience

BACKGROUND

The Greulich and Pyle (GP) method is one of the most common radiographic techniques for bone age (BA) assessment. The applicability of this method to ethnic populations outside of the United States has been investigated in several recent studies worldwide. Currently, limited data are available on the accuracy of the GP method for the Taiwanese population. The purpose of this study was to determine whether the GP standards are applicable to contemporary Taipei children.

METHODS

Clinical data from October 1, 2010, to March 31, 2020, were retrospectively collected from a general hospital in Taipei. BA was determined by a senior pediatrician and was reviewed by a senior pediatric radiologist according to the GP standards. Comparison of BA and chronological age (CA) was performed in children with body weight and height in the 15th to the 85th percentiles of normal children. Ethnic variations in the maturation process in the ulnar bone were investigated. All data were statistically analyzed.

RESULTS

In total, 2465 medical records were collected. After excluding those with diseases and unqualified data, 654 records of boys and 809 of girls were analyzed. In boys, the mean BA was significantly delayed between 6 and 9 years of age compared with the CA. In girls, the mean BA was generally advanced between 7 and 15 years of age. Ulnar bone maturation tended to be delayed in young boys.

CONCLUSION

A significant discrepancy between CA and BA was observed in our population. Delayed ulnar bone maturation in young boys was confirmed. Children in Taipei exhibit a different maturation pattern than children on whom the GP standards were based.

Study of Multidimensional and High-Precision Height Model of Youth Based on Multilayer Perceptron

Predicting the adult height of children accurately has great social value for the selection of outstanding athlete as well as early detection of children's growth disorders. Currently, the mainstream method used to predict adult height in China has three problems: its standards are not uniform; it is stale for current Chinese children; its accuracy is not satisfactory. This article uses the data collected by the Chinese Children and Adolescents' Physical Fitness and Growth Health Project in Zhejiang primary and secondary schools. We put forward a new multidimensional and high-precision youth growth curve prediction model, which is based on multilayer perceptron. First, this model uses multidimensional growth data of children as predictors and then utilizes multilayer perceptron to predict the children's adult height. Second, we find the Table of Height Standard Deviation of Chinese Children and fit the data of zero standard deviation to obtain the curve. This curve is regarded as Chinese children's mean growth curve. Third, we use the least-squares method and the mean curve to calculate the individual growth curve. Finally, the individual curve can be used to predict children's state height. Experimental results show that this adult height prediction model's accuracy (between 2 cm) of boys and girls reached 90.20% and 88.89% and the state height prediction accuracy reached 77.46% and 74.93%. Compared with Bayley–Pinneau, the adult height prediction is improved 19.61% for boys and 13.33% for girls. Compared with BoneXpert, the adult height prediction is improved 25.49% for boys and 6.67% for girls. Compared with the method based on the bone age growth map, the adult height prediction is improved 15.69% for boys and 24.45% for girls.

Reference Curves for Pediatric Endocrinology: Leveraging Biomarker Z-Scores for Clinical Classifications

CONTEXT

Hormone reference intervals in pediatric endocrinology are traditionally partitioned by age and lack the framework for benchmarking individual blood test results as normalized z-scores and plotting sequential measurements onto a chart. Reference curve modeling is applicable to endocrine variables and represents a standardized method to account for variation with gender and age.

OBJECTIVE

We aimed to establish gender-specific biomarker reference curves for clinical use and benchmark associations between hormones, pubertal phenotype, and body mass index (BMI).

METHODS

Using cross-sectional population sample data from 2139 healthy Norwegian children and adolescents, we analyzed the pubertal status, ultrasound measures of glandular breast tissue (girls) and testicular volume (boys), BMI, and laboratory measurements of 17 clinical biomarkers modeled using the established "LMS" growth chart algorithm in R.

RESULTS

Reference curves for puberty hormones and pertinent biomarkers were modeled to adjust for age and gender. Z-score equivalents of biomarker levels and anthropometric measurements were compiled in a comprehensive beta coefficient matrix for each gender. Excerpted from this analysis and independently of age, BMI was positively associated with female glandular breast volume (β = 0.5, P < 0.001) and leptin (β = 0.6, P < 0.001), and inversely correlated with serum levels of sex hormone-binding globulin (SHBG) (β = -0.4, P < 0.001). Biomarker z-score profiles differed significantly between cohort subgroups stratified by puberty phenotype and BMI weight class.

CONCLUSION

Biomarker reference curves and corresponding z-scores provide an intuitive framework for clinical implementation in pediatric endocrinology and facilitate the application of machine learning classification and covariate precision medicine for pediatric patients.

Heterozygous NPR2 Variants in Idiopathic Short Stature

Heterozygous variants in the NPR2 gene, which encodes the B-type natriuretic peptide receptor (NPR-B), a regulator of skeletal growth, were reported in 2-6% cases of idiopathic short stature (ISS). Using next-generation sequencing (NGS), we aimed to assess the frequency of NPR2 variants in our study cohort consisting of 150 children and adolescents with ISS, describe the NPR2 phenotypic spectrum with a growth pattern including birth data, and study the response to growth hormone (GH) treatment. A total of ten heterozygous pathogenic/likely pathogenic NPR2 variants and two heterozygous NPR2 variants of uncertain significance were detected in twelve participants (frequency of causal variants: 10/150, 6.7%). During follow-up, the NPR2 individuals presented with a growth pattern varying from low-normal to significant short stature. A clinically relevant increase in BMI (a mean gain in the BMI SDS of +1.41), a characteristic previously not reported in NPR2 individuals, was observed. In total, 8.8% participants born small for their gestational age (SGA) carried the NPR2 causal variant. The response to GH treatment was variable (SDS height gain ranging from -0.01 to +0.74). According to the results, NPR2 variants present a frequent cause of ISS and familial short stature. Phenotyping variability in growth patterns and variable responses to GH treatment should be considered.

Validation of automated bone age analysis from hand radiographs in a North American pediatric population

BACKGROUND

Radiographic bone age assessment by automated software is precise and instantaneous.

OBJECTIVE

The aim of this study was to evaluate the accuracy of an automated tool for bone age assessment.

MATERIALS AND METHODS

We compared a total of 586 bone age radiographs from 451 patients, which had been assessed by three radiologists from 2013 to 2018, with bone age analysis by BoneXpert, using the Greulich and Pyle method. We made bone age comparisons in different patient groups based on gender, diagnosis and race, and in a subset with repeated bone age studies. We calculated Spearman correlation (r) and accuracy (root mean square error, or R²).

RESULTS

Bone age analyses by automated and manual assessments showed a strong correlation (r=0.98; R²=0.96; P<0.0001), with the mean bone age difference of 0.12±0.76 years. Bone age comparisons by the two methods remained strongly correlated (P<0.0001) when stratified by gender, common endocrine conditions including growth disorders and early/precocious puberty, and race. In the longitudinal analysis, we also found a strong correlation between the automated software and manual bone age over time (r=0.7852; R²=0.63; P<0.01).

CONCLUSION

Automated bone age assessment was found to be reliable and accurate in a large cohort of pediatric patients in a clinical practice setting in North America.

Accuracy and self-validation of automated bone age determination

The BoneXpert method for automated determination of bone age from hand X-rays was introduced in 2009 and is currently running in over 200 hospitals. The aim of this work is to present version 3 of the method and validate its accuracy and self-validation mechanism that automatically rejects an image if it is at risk of being analysed incorrectly. The training set included 14,036 images from the 2017 Radiological Society of North America (RSNA) Bone Age Challenge, 1642 images of normal Dutch and Californian children, and 8250 images from Tübingen from patients with Short Stature, Congenital Adrenal Hyperplasia and Precocious Puberty. The study resulted in a cross-validated root mean square (RMS) error in the Tübingen images of 0.62 y, compared to 0.72 y in the previous version. The RMS error on the RSNA test set of 200 images was 0.45 y relative to the average of six manual ratings. The self-validation mechanism rejected 0.4% of the RSNA images. 121 outliers among the self-validated images of the Tübingen study were rerated, resulting in 6 cases where BoneXpert deviated more than 1.5 years from the average of the three re-ratings, compared to 72 such cases for the original manual ratings. The accuracy of BoneXpert is clearly better than the accuracy of a single manual rating. The self-validation mechanism rejected very few images, typically with abnormal anatomy, and among the accepted images, there were 12 times fewer severe bone age errors than in manual ratings, suggesting that BoneXpert could be safer than manual rating.

External validation of deep learning-based bone-age software: a preliminary study with real world data

Artificial intelligence (AI) is increasingly being used in bone-age (BA) assessment due to its complicated and lengthy nature. We aimed to evaluate the clinical performance of a commercially available deep learning (DL)–based software for BA assessment using a real-world data. From Nov. 2018 to Feb. 2019, 474 children (35 boys, 439 girls, age 4–17 years) were enrolled. We compared the BA estimated by DL software (DL-BA) with that independently estimated by 3 reviewers (R1: Musculoskeletal radiologist, R2: Radiology resident, R3: Pediatric endocrinologist) using the traditional Greulich–Pyle atlas, then to his/her chronological age (CA). A paired t-test, Pearson’s correlation coefficient, Bland–Altman plot, mean absolute error (MAE) and root mean square error (RMSE) were used for the statistical analysis. The intraclass correlation coefficient (ICC) was used for inter-rater variation. There were significant differences between DL-BA and each reviewer’s BA (P < 0.025), but the correlation was good with one another (r = 0.983, P < 0.025). RMSE (MAE) values were 10.09 (7.21), 10.76 (7.88) and 13.06 (10.06) months between DL-BA and R1, R2, R3 BA. Compared with the CA, RMSE (MAE) values were 13.54 (11.06), 15.18 (12.11), 16.19 (12.78) and 19.53 (17.71) months for DL-BA, R1, R2, R3 BA, respectively. Bland–Altman plots revealed the software and reviewers’ tendency to overestimate the BA in general. ICC values between 3 reviewers were 0.97, 0.85 and 0.86, and the overall ICC value was 0.93. The BA estimated by DL-based software showed statistically similar, or even better performance than that of reviewers’ compared to the chronological age in the real world clinic.

Final adult height in children with central precocious puberty - a retrospective study

BACKGROUND/AIMS

Central precocious puberty (CPP) is due to premature activation of the hypothalamic-pituitary-gonadal axis. It predominantly affects girls. CPP leads to lower final height (FH), yet the treatment benefit in girls between 6 and 8 years is equivocal. Our main goal was to evaluate the effects of gonadotropin-releasing hormone analog (GnRHa) on FH and identify factors that predict FH.

METHODS

In a retrospective study, children with CPP (12 boys, 81 girls) that reached FH were included. Their clinical data at diagnosis and up to their final height was compared by descriptive statistics among idiopathic (iCPP) (n=68) and non-idiopathic CPP (nCPP) and between GnRHa treated (n=48) and untreated (n=15) girls with iCPP. The treatment effect of body weight (BW) adjusted GnRHa dosing was evaluated. Univariate linear regression and step-wise multivariable regression including 48 girls with iCPP treated with GnRHa were performed to identify predicting factors for FH.

RESULTS

Children with idiopathic CPP (iCPP) reached higher FH (p=0.002) than children with non-idiopathic CPP. After the diagnosis, the treated group gained 7.0 cm more than the untreated group. Yet, attributable to individualized decision-making, the FH in both groups was comparable (161.5 cm in treated, 161.0 cm in untreated girls with iCPP), although the onset of menarche was 2.5 years earlier among untreated girls. BW-adjusted dosing suppressed peak luteinizing hormone (LH) below 4.5 IU/L in 95% of children; however, bone age further advanced during therapy in 38% of patients. Predicting factors revealed by multivariable regression were bone age at diagnosis, BMI SDS at diagnosis, LH basal, age at start and cessation of treatment, predicted adult height and target height. (R2 = 0.72).

CONCLUSION

Children with nCPP had worse FH outcome compared to iCPP despite similar CPP onset and therapeutic characteristics. Treatment by GnRHa using BW-adjusted dosing was effective in delaying menarche onset and reaching target height in girls with iCPP. Multiple factors affecting FH outcome indicated individualized decision-making regarding therapeutic intervention remains challenging. In the treated patients, among the factors that can be influenced, height at treatment cessation most significantly influenced the outcome.

Factors affecting prepubertal and pubertal bone age progression

Bone age (BA) is a clinical marker of bone maturation which indicates the developmental stage of endochondral ossification at the epiphysis and the growth plate. Hormones that promote the endochondral ossification process include growth hormone, insulin-like growth factor-1, thyroid hormone, estrogens, and androgens. In particular, estrogens are essential for growth plate fusion and closure in both sexes. Bone maturation in female children is more advanced than in male children of all ages. The promotion of bone maturation seen in females before the onset of puberty is thought to be an effect of estrogen because estrogen levels are higher in females than in males before puberty. Sex hormones are essential for bone maturation during puberty. Since females have their pubertal onset about two years earlier than males, bone maturation in females is more advanced than in males during puberty. In the present study, we aimed to review the factors affecting prepubertal and pubertal BA progression, BA progression in children with hypogonadism, and bone maturation and deformities in children with Turner syndrome.

An overview of deep learning applications in precocious puberty and thyroid dysfunction

In the last decade, deep learning methods have garnered a great deal of attention in endocrinology research. In this article, we provide a summary of current deep learning applications in endocrine disorders caused by either precocious onset of adult hormone or abnormal amount of hormone production. To give access to the broader audience, we start with a gentle introduction to deep learning and its most commonly used architectures, and then we focus on the research trends of deep learning applications in thyroid dysfunction classification and precocious puberty diagnosis. We highlight the strengths and weaknesses of various approaches and discuss potential solutions to different challenges. We also go through the practical considerations useful for choosing (and building) the deep learning model, as well as for understanding the thought process behind different decisions made by these models. Finally, we give concluding remarks and future directions.

Autonomous artificial intelligence in pediatric radiology: the use and perception of BoneXpert for bone age assessment

BACKGROUND

The autonomous artificial intelligence (AI) system for bone age rating (BoneXpert) was designed to be used in clinical radiology practice as an AI-replace tool, replacing the radiologist completely.

OBJECTIVE

The aim of this study was to investigate how the tool is used in clinical practice. Are radiologists more inclined to use BoneXpert to assist rather than replace themselves, and how much time is saved?

MATERIALS AND METHODS

We sent a survey consisting of eight multiple-choice questions to 282 radiologists in departments in Europe already using the software.

RESULTS

The 97 (34%) respondents came from 18 countries. Their answers revealed that before installing the automated method, 83 (86%) of the respondents took more than 2 min per bone age rating; this fell to 20 (21%) respondents after installation. Only 17/97 (18%) respondents used BoneXpert to completely replace the radiologist; the rest used it to assist radiologists to varying degrees. For instance, 39/97 (40%) never overruled the automated reading, while 9/97 (9%) overruled more than 5% of the automated ratings. The majority 58/97 (60%) of respondents checked the radiographs themselves to exclude features of underlying disease.

CONCLUSION

BoneXpert significantly reduces reporting times for bone age determination. However, radiographic analysis involves more than just determining bone age. It also involves identification of abnormalities, and for this reason, radiologists cannot be completely replaced. AI systems originally developed to replace the radiologist might be more suitable as AI assist tools, particularly if they have not been validated to work autonomously, including the ability to omit ratings when the image is outside the range of validity.

Identifying Skeletal Maturity from X-rays using Deep Neural Networks

Skeletal maturity estimation is routinely evaluated by pediatrics and radiologists to assess growth and hormonal disorders. Methods integrated with regression techniques are incompatible with low-resolution digital samples and generate bias, when the evaluation protocols are implemented for feature assessment on coarse X-Ray hand images. This paper proposes a comparative analysis between two deep neural network architectures, with the base models such as Inception-ResNet-V2 and Xception-pre-trained networks. Based on 12,611 hand X-Ray images of RSNA Bone Age database, Inception-ResNet-V2 and Xception models have achieved R-Squared value of 0.935 and 0.942 respectively. Further, in the same order, the MAE accomplished by the two models are 12.583 and 13.299 respectively, when subjected to very few training instances with negligible chances of overfitting.

Tanner-Whitehouse and Modified Bayley-Pinneau Adult Height Predictions in Elite Youth Soccer Players from the Middle East

PURPOSE

To provide the first scrutiny of adult height prediction protocols based on automated Greulich-Pyle and Tanner-Whitehouse (TW) skeletal ages applied to elite youth soccer players from the Middle East.

METHODS

We examined the application of modified Bayley-Pinneau (BoneXpert®), TW-II, and TW-III protocols using mixed-longitudinal data available for 103 subjects (chronological age range, 19.4 to 27.9 yr) previously enrolled as academy student-athletes (annual screening range, one to six visits). Random-effects generalized additive models quantified the presence of systematic mean differences between actual versus predicted adult height. Effects were deemed practically equivalent based on the location of the confidence interval (95% CI) against a realistic difference value of Δ = ± 1 cm. Each model pooled residual standard deviation described the actual precision of height predictions and was used to calculate a 95% prediction interval.

RESULTS

The BoneXpert® method overpredicted adult height systematically at chronological ages in the range of approximately 13.5 to 14.5 yr (95% CI range, -1.9 to -1 cm) and Greulich-Pyle skeletal ages between 13.5 and 15 yr (95% CI range, -3.1 to -1 cm). Effects based on TW-II were practically equivalent across the chronological and skeletal age measurement ranges, with this protocol yielding adult height predictions with a precision (standard deviation) of approximately ±2.6 cm. The mean TW-III effects indicated systematic adult height overpredictions until the attainment of 14.5 and 15 yr of chronological age (95% CI range, -3.8 to -1.1 cm) and TW-III skeletal age (95% CI range: -5.2 to -2.3 cm), respectively.

CONCLUSIONS

Tanner-Whitehouse-II adult height prediction method provided more consistent estimates and can be considered the method of choice for talent development purposes in youth soccer players from the Middle East.

Clinical Validation of a Deep Learning-Based Hybrid (Greulich-Pyle and Modified Tanner-Whitehouse) Method for Bone Age Assessment

OBJECTIVE

To evaluate the accuracy and clinical efficacy of a hybrid Greulich-Pyle (GP) and modified Tanner-Whitehouse (TW) artificial intelligence (AI) model for bone age assessment.

MATERIALS AND METHODS

A deep learning-based model was trained on an open dataset of multiple ethnicities. A total of 102 hand radiographs (51 male and 51 female; mean age ± standard deviation = 10.95 ± 2.37 years) from a single institution were selected for external validation. Three human experts performed bone age assessments based on the GP atlas to develop a reference standard. Two study radiologists performed bone age assessments with and without AI model assistance in two separate sessions, for which the reading time was recorded. The performance of the AI software was assessed by comparing the mean absolute difference between the AI-calculated bone age and the reference standard. The reading time was compared between reading with and without AI using a paired t test. Furthermore, the reliability between the two study radiologists' bone age assessments was assessed using intraclass correlation coefficients (ICCs), and the results were compared between reading with and without AI.

RESULTS

The bone ages assessed by the experts and the AI model were not significantly different (11.39 ± 2.74 years and 11.35 ± 2.76 years, respectively, p = 0.31). The mean absolute difference was 0.39 years (95% confidence interval, 0.33-0.45 years) between the automated AI assessment and the reference standard. The mean reading time of the two study radiologists was reduced from 54.29 to 35.37 seconds with AI model assistance (p < 0.001). The ICC of the two study radiologists slightly increased with AI model assistance (from 0.945 to 0.990).

CONCLUSION

The proposed AI model was accurate for assessing bone age. Furthermore, this model appeared to enhance the clinical efficacy by reducing the reading time and improving the inter-observer reliability.

Pediatric Pulmonary Embolism: Imaging Guidelines and Recommendations

In contrast with the algorithms and screening criteria available for adults with suspected pulmonary embolism, there is a paucity of guidance on the diagnostic approach for children. The incidence of pulmonary embolism in the pediatric population and young adults is higher than thought, and there is an urgent need for updated guidelines for the imaging approach to diagnosis in the pediatric population. This article presents an up-to-date review of imaging techniques, characteristic radiologic findings, and an evidence-based algorithm for the detection of pediatric pulmonary embolism to improve the care of pediatric patients with suspected pulmonary embolism.

Generative adversarial network based data augmentation and gender-last training strategy with application to bone age assessment

BACKGROUND AND OBJECTIVES

Bone age assessment (BAA) is widely used in determination of discrepancy between skeletal age and chronological age. Manual approaches are complicated which require experienced experts, while existing automatic approaches are perplexed with small and imbalanced samples which is a big challenge in deep learning.

METHODS

In this study, we proposed a new deep learning based method to improve the BAA training in both pre-training and training architecture. In pre-training, we proposed a framework using a new distance metric of cosine distance in the framework of optimal transport for data augmentation (CNN-GAN-OTD). In the training architecture, we explored the order of gender label and bone age information, supervised and semi-supervised training.

RESULTS

We found that the training architecture with the CNN-GAN-OTD based data augmentation and supervised gender-last classification with supervised Inception v3 network yielded the best assessment (mean average error of 4.23 months).

CONCLUSIONS

The proposed data augmentation framework could be a potential built-in component of general deep learning networks and the training strategy with different label order could inspire more and deeper consideration of label priority in multi-label tasks.

Data Enhancement and Deep Learning for Bone Age Assessment using The Standards of Skeletal Maturity of Hand and Wrist for Chinese

Conventional methods for artificial age determination of skeletal bones have several problems, such as strong subjectivity, large random errors, complex evaluation processes, and long evaluation cycles. In this study, an automated age determination of skeletal bones was performed based on Deep Learning. Two methods were used to evaluate bone age, one based on examining all bones in the palm and another based on the deep convolutional neural network (CNN) method. Both methods were evaluated using the same test dataset. Moreover, we can extend the dataset and increase the generalisation ability of the network by data expansion. Consequently, a more accurate bone age can be obtained. This method can reduce the average error of the final bone age evaluation and lower the upper limit of the absolute value of the error of the single bone age. The experiments show the effectiveness of the proposed method, which can provide doctors and users with more stable, efficient and convenient diagnosis support and decision support.

Deep Learning Framework for Automatic Bone Age Assessment

Bone age Assessment or the skeletal age is a general clinical practice to detect endocrine and metabolic disarrangement in child development. The bone age indicates the level of structural and biological growth better than chronological age calculated from the birth date. The X-Ray of the wrist and hand is used in common to estimate the bone age of a person. The degree of agreement among the automated methods used to evaluate the X-rays is more than any other manual method. In this work, we propose a fully automated deep learning approach for bone age assessment. The dataset used is from the 2017 Pediatric Bone Age Challenge released by the Radiological Society of North America. Each X-Ray image in this dataset is an image of a left hand tagged with the age and gender of the patient. Transfer learning is employed by using pre-trained neural network architecture. InceptionV3 architecture is used in the present work, and the difference between the actual and predicted age obtained is 5.921 months.Clinical Relevance- This provides an AI-based computer assistance system as a supplement tool to help clinicians make bone age predictions.

Self-Supervised Attention Mechanism for Pediatric Bone Age Assessment With Efficient Weak Annotation

Pediatric bone age assessment (BAA) is a common clinical practice to investigate endocrinology, genetic and growth disorders of children. Different specific bone parts are extracted as anatomical Regions of Interest (RoIs) during this task, since their morphological characters have important biological identification in skeletal maturity. Following this clinical prior knowledge, recently developed deep learning methods address BAA with an RoI-based attention mechanism, which segments or detects the discriminative RoIs for meticulous analysis. Great strides have been made, however, these methods strictly require large and precise RoIs annotations, which limits the real-world clinical value. To overcome the severe requirements on RoIs annotations, in this paper, we propose a novel self-supervised learning mechanism to effectively discover the informative RoIs without the need of extra knowledge and precise annotation-only image-level weak annotation is all we take. Our model, termed PEAR-Net for Part Extracting and Age Recognition Network, consists of one Part Extracting (PE) agent for discriminative RoIs discovering and one Age Recognition (AR) agent for age assessment. Without precise supervision, the PE agent is designed to discover and extract RoIs fully automatically. Then the proposed RoIs are fed into AR agent for feature learning and age recognition. Furthermore, we utilize the self-consistency of RoIs to optimize PE agent to understand the part relation and select the most useful RoIs. With this self-supervised design, the PE agent and AR agent can reinforce each other mutually. To the best of our knowledge, this is the first end-to-end bone age assessment method which can discover RoIs automatically with only image-level annotation. We conduct extensive experiments on the public RSNA 2017 dataset and achieve state-of-the-art performance with MAE 3.99 months. Project is available at http://imcc.ustc.edu.cn/project/ssambaa/.

Detection of Del/Dup Inside SHOX/PAR1 Region in Children and Young Adults with Idiopathic Short Stature

Short stature is a common growth disorder defined as a body height two standard deviations (SD) or more below the mean for a given age, gender, and population. A large part of the cases remains unexplained and is referred to as having idiopathic short stature (ISS). One of the leading genetic causes of short stature is variants of short stature homeobox-containing gene (SHOX) and is considered to be responsible for 2–15% of ISS. We aimed to analyse the regulatory and coding region of SHOX in Slovenian children and young adults with ISS and to investigate the pathogenicity of detected variants. Our cohort included 75 children and young adults with ISS. Multiplex ligation-dependent probe amplification (MLPA) was performed in all participants for the detection of larger copy number variations (CNVs). Sanger sequencing was undertaken for the detection of point variants, small deletions, and insertions. A total of one deletion and two duplications were discovered using the MLPA technique. Only one of these four variants was identified as disease-causing and occurred in one individual, which represents 1.3% of the cohort. With Sanger sequencing, two variants were discovered, but none of them appeared to have a pathogenic effect on height. According to the results, in the Slovenian population of children and young adults with ISS, SHOX deficiency is less frequent than expected considering existing data from other populations.

Using Deep Neural Networks for Predicting Age and Sex in Healthy Adult Chest Radiographs

Background: The performance of chest radiography-based age and sex prediction has not been well validated. We used a deep learning model to predict the age and sex of healthy adults based on chest radiographs (CXRs). Methods: In this retrospective study, 66,643 CXRs of 47,060 healthy adults were used for model training and testing. In total, 47,060 individuals (mean age ± standard deviation, 38.7 ± 11.9 years; 22,144 males) were included. By using chronological ages as references, mean absolute error (MAE), root mean square error (RMSE), and Pearson’s correlation coefficient were used to assess the model performance. Summarized class activation maps were used to highlight the activated anatomical regions. The area under the curve (AUC) was used to examine the validity for sex prediction. Results: When model predictions were compared with the chronological ages, the MAE was 2.1 years, RMSE was 2.8 years, and Pearson’s correlation coefficient was 0.97 (p < 0.001). Cervical, thoracic spines, first ribs, aortic arch, heart, rib cage, and soft tissue of thorax and flank seemed to be the most crucial activated regions in the age prediction model. The sex prediction model demonstrated an AUC of >0.99. Conclusion: Deep learning can accurately estimate age and sex based on CXRs.

Traditional and New Methods of Bone Age Assessment-An Overview

Bone age is one of biological indicators of maturity used in clinical practice and it is a very important parameter of a child’s assessment, especially in paediatric endocrinology. The most widely used method of bone age assessment is by performing a hand and wrist radiograph and its analysis with Greulich-Pyle or Tanner-Whitehouse atlases, although it has been about 60 years since they were published. Due to the progress in the area of Computer-Aided Diagnosis and application of artificial intelligence in medicine, lately, numerous programs for automatic bone age assessment have been created. Most of them have been verified in clinical studies in comparison to traditional methods, showing good precision while eliminating inter- and intra-rater variability and significantly reducing the time of assessment. Additionally, there are available methods for assessment of bone age which avoid X-ray exposure, using modalities such as ultrasound or magnetic resonance imaging.

Congenital Adrenal Hyperplasia in Children: A Pilot Study of Steroid Hormones Expressed as Sex- and Age-Related Standard Deviation Scores

INTRODUCTION

Congenital adrenal hyperplasia (CAH) is an autosomal recessive disease predominantly caused by 21-hydroxylase deficiency. Clinical management in children includes glucocorticoid and often mineralocorticoid treatment alongside monitoring outcomes such as an-thro-po-metry, pubertal status, blood pressure, and biochemistry.

OBJECTIVE

The objective of this pilot study was to present the use of 17-hydroxyprogesterone (17-OHP) and androgen metabolites expressed as standard deviation (SD) scores rather than actual concentrations as a tool in the management of children with CAH as well as in research settings.

METHODS

The study was a retrospective, longitudinal study that took place in a single, tertiary center and included 38 children and adolescents aged 3-18 years with CAH due to 21-hydroxylase deficiency. Biochemical measurements of 17-OHP, androstenedione, dehydroepiandrosterone-sulphate (DHEAS), and testosterone using liquid chromatography-tandem mass spectrometry were expressed as SD scores, and outcomes such as genotype, height, bone maturation, blood pressure, and treatment doses were extracted from patient files.

RESULTS

The majority (86%) of CAH patients had 17-OHP measurements above +2 SD during standard hydrocortisone therapy, receiving an average daily hydrocortisone dose of 12.6 mg/m2. Androstenedione concentrations were mostly within ±2 SD, whereas DHEAS values were below -2 SD in 47% of patients.

CONCLUSIONS

Applying sex- and age-related SD scores to 17-OHP and androgen metabolite concentrations allows for monitoring of hydrocortisone treatment independent of age, sex, assay, and center. We propose that 17-OHP and androgen metabolites expressed as SD scores be implemented as a unifying tool that simplifies research and, in the future, also optimal management of treatment.

Familial central precocious puberty: two novel MKRN3 mutations

BACKGROUND

Paternally inherited loss-of-function mutations in MKRN3 underlie central precocious puberty (CPP). We describe clinical and genetic features of CPP patients with paternally inherited MKRN3 mutations in two independent families.

METHODS

The single coding exon of MKRN3 was analyzed in three patients with CPP and their family members, followed by segregation analyses. Additionally, we report the patients' responses to GnRH analog treatment.

RESULTS

A paternally inherited novel heterozygous c.939C>G, p.(Ile313Met) missense mutation affecting the RING finger domain of MKRN3 was found in a Finnish girl with CPP (age at presentation 6 years). Two Polish siblings (a girl presenting with B2 at the age of 4 years and a boy with adult size testes at the age of 9 years) had inherited a novel heterozygous MKRN3 mutation c.1237_1252delGGAGACACATGCTTTT p.(Gly413Thrfs*63) from their father. The girls were treated with GnRH analogs, which exhibited suppression of the hypothalamic-pituitary-gonadal axis. In contrast, the male patient was not treated, yet he reached his target height.

CONCLUSIONS

We describe two novel MKRN3 mutations in three CPP patients. The first long-term data on a boy with CPP due to an MKRN3 mutation questions the role of GnRH analog treatment in augmenting adult height in males with this condition.

IMPACT

We describe the genetic cause for central precocious puberty (CPP) in two families. This report adds two novel MKRN3 mutations to the existing literature. One of the mutations, p.(Ile313Met) affects the RING finger domain of MKRN3, which has been shown to be important for repressing the promoter activity of KISS1 and TAC3. We describe the first long-term observation of a male patient with CPP due to a paternally inherited MKRN3 loss-of-function mutation. Without GnRH analog treatment, he achieved an adult height that was in accordance with his mid-parental target height.

Attention-Guided Discriminative Region Localization and Label Distribution Learning for Bone Age Assessment

Bone age assessment (BAA) is clinically important as it can be used to diagnose endocrine and metabolic disorders during child development. Existing deep learning based methods for classifying bone age use the global image as input, or exploit local information by annotating extra bounding boxes or key points. However, training with the global image underutilizes discriminative local information, while providing extra annotations is expensive and subjective. In this paper, we propose an attention-guided approach to automatically localize the discriminative regions for BAA without any extra annotations. Specifically, we first train a classification model to learn the attention maps of the discriminative regions, finding the hand region, the most discriminative region (the carpal bones), and the next most discriminative region (the metacarpal bones). Guided by those attention maps, we then crop the informative local regions from the original image and aggregate different regions for BAA. Instead of taking BAA as a general regression task, which is suboptimal due to the label ambiguity problem in the age label space, we propose using joint age distribution learning and expectation regression, which makes use of the ordinal relationship among hand images with different individual ages and leads to more robust age estimation. Extensive experiments are conducted on the RSNA pediatric bone age data set. {\color{red} Without using extra manual} annotations, our method achieves competitive results compared with existing state-of-the-art deep learning-based methods that require manual annotation. Code is available at \url{https://github.com/chenchao666/Bone-Age-Assessment}.

Bone structure assessed with pQCT in prepubertal males with delayed puberty or congenital hypogonadotropic hypogonadism

OBJECTIVE

Congenital hypogonadotropic hypogonadism (CHH) is associated with impaired bone mineral density in adulthood, whereas the estimates on bone structure in adolescents with CHH has not been previously evaluated. This study describes bone structure in CHH patients and compares it to that in boys with constitutional delay of growth and puberty (CDGP).

DESIGN

A cross-sectional study.

METHODS

Peripheral quantitative computed tomography (pQCT) of non-dominant arm and left leg were performed. Volumetric bone mineral density (BMD), bone mineral content, and area in trabecular and cortical bone compartments were evaluated, and bone age-adjusted Z-scores for the bone parameters were determined.

RESULTS

The participants with CHH had more advanced bone age and were older, taller and heavier than the CDGP boys, yet they had lower trabecular BMD in distal radius (147.7 mg/mm³ [95% CI, 128-168 mg/mm³ ] vs. 181.2 mg/mm³ [172-192 mg/mm³ ], p = .002) and distal tibia (167.6 mg/mm³ [145-190 mg/mm³ ] vs. 207.2 mg/mm³ [187-227 mg/mm³ ], p = .012), respectively. CHH males had lower cortical thickness at diaphyseal tibia than the participants with CDGP (p = .001). These between-group differences remained significant in corresponding Z-scores adjusted for bone age and height (p = .001). In CDGP group, serum testosterone correlated positively with trabecular BMD (r = 0.51, p = .013) at distal radius, and estradiol levels correlated positively with trabecular BMD at the distal site of tibia (r = 0.58, p = .004).

CONCLUSIONS

Five treatment-naïve male patients with CHH exhibited poorer trabecular BMD than untreated males with CDGP. We speculate that timely low-dose sex steroid replacement in CHH males may benefit skeletal health in adulthood.

Assessment of rapidly advancing bone age during puberty on elbow radiographs using a deep neural network model

OBJECTIVES

Bone age is considered an indicator for the diagnosis of precocious or delayed puberty and a predictor of adult height. We aimed to evaluate the performance of a deep neural network model in assessing rapidly advancing bone age during puberty using elbow radiographs.

METHODS

In all, 4437 anteroposterior and lateral pairs of elbow radiographs were obtained from pubertal individuals from two institutions to implement and validate a deep neural network model. The reference standard bone age was established by five trained researchers using the Sauvegrain method, a scoring system based on the shapes of the lateral condyle, trochlea, olecranon apophysis, and proximal radial epiphysis. A test set (n = 141) was obtained from an external institution. The differences between the assessment of the model and that of reviewers were compared.

RESULTS

The mean absolute difference (MAD) in bone age estimation between the model and reviewers was 0.15 years on internal validation. In the test set, the MAD between the model and the five experts ranged from 0.19 to 0.30 years. Compared with the reference standard, the MAD was 0.22 years. Interobserver agreement was excellent among reviewers (ICC: 0.99) and between the model and the reviewers (ICC: 0.98). In the subpart analysis, the olecranon apophysis exhibited the highest accuracy (74.5%), followed by the trochlea (73.7%), lateral condyle (73.7%), and radial epiphysis (63.1%).

CONCLUSIONS

Assessment of rapidly advancing bone age during puberty on elbow radiographs using our deep neural network model was similar to that of experts.

KEY POINTS

• Bone age during puberty is particularly important for patients with scoliosis or limb-length discrepancy to determine the phase of the disease, which influences the timing and method of surgery. • The commonly used hand radiographs-based methods have limitations in assessing bone age during puberty due to the less prominent morphological changes of the hand and wrist bones in this period. • A deep neural network model trained with elbow radiographs exhibited similar performance to human experts on estimating rapidly advancing bone age during puberty.

Important Tools for Use by Pediatric Endocrinologists in the Assessment of Short Stature

Assessment and management of children with growth failure has improved greatly over recent years. However, there remains a strong potential for further improvements by using novel digital techniques. A panel of experts discussed developments in digitalization of a number of important tools used by pediatric endocrinologists at the third 360° European Meeting on Growth and Endocrine Disorders, funded by Merck KGaA, Germany, and this review is based on those discussions. It was reported that electronic monitoring and new algorithms have been devised that are providing more sensitive referral for short stature. In addition, computer programs have improved ways in which diagnoses are coded for use by various groups including healthcare providers and government health systems. Innovative cranial imaging techniques have been devised that are considered safer than using gadolinium contrast agents and are also more sensitive and accurate. Deep-learning neural networks are changing the way that bone age and bone health are assessed, which are more objective than standard methodologies. Models for prediction of growth response to growth hormone (GH) treatment are being improved by applying novel artificial intelligence methods that can identify non-linear and linear factors that relate to response, providing more accurate predictions. Determination and interpretation of insulin-like growth factor-1 (IGF-1) levels are becoming more standardized and consistent, for evaluation across different patient groups, and computer-learning models indicate that baseline IGF-1 standard deviation score is among the most important indicators of GH therapy response. While physicians involved in child growth and treatment of disorders resulting in growth failure need to be aware of, and keep abreast of, these latest developments, treatment decisions and management should continue to be based on clinical decisions. New digital technologies and advancements in the field should be aimed at improving clinical decisions, making greater standardization of assessment and facilitating patient-centered approaches.

Probing an AI regression model for hand bone age determination using gradient-based saliency mapping

Understanding how a neural network makes decisions holds significant value for users. For this reason, gradient-based saliency mapping was tested on an artificial intelligence (AI) regression model for determining hand bone age from X-ray radiographs. The partial derivative (PD) of the inferred age with respect to input image intensity at each pixel served as a saliency marker to find sensitive areas contributing to the outcome. The mean of the absolute PD values was calculated for five anatomical regions of interest, and one hundred test images were evaluated with this procedure. The PD maps suggested that the AI model employed a holistic approach in determining hand bone age, with the wrist area being the most important at early ages. However, this importance decreased with increasing age. The middle section of the metacarpal bones was the least important area for bone age determination. The muscular region between the first and second metacarpal bones also exhibited high PD values but contained no bone age information, suggesting a region of vulnerability in age determination. An end-to-end gradient-based saliency map can be obtained from a black box regression AI model and provide insight into how the model makes decisions.

Automated Bone Age Assessment Using Artificial Intelligence: The Future of Bone Age Assessment

Bone age assessments are a complicated and lengthy process, which are prone to inter- and intra-observer variabilities. Despite the great demand for fully automated systems, developing an accurate and robust bone age assessment solution has remained challenging. The rapidly evolving deep learning technology has shown promising results in automated bone age assessment. In this review article, we will provide information regarding the history of automated bone age assessments, discuss the current status, and present a literature review, as well as the future directions of artificial intelligence-based bone age assessments.

A deep learning-based computer-aided diagnosis method of X-ray images for bone age assessment

Bone age assessment using hand-wrist X-ray images is fundamental when diagnosing growth disorders of a child or providing a more patient-specific treatment. However, as clinical procedures are a subjective assessment, the accuracy depends highly on the doctor’s experience. Motivated by this, a deep learning-based computer-aided diagnosis method was proposed for performing bone age assessment. Inspired by clinical approaches and aimed to reduce expensive manual annotations, informative regions localization based on a complete unsupervised learning method was firstly performed and an image-processing pipeline was proposed. Subsequently, an image model with pre-trained weights as a backbone was utilized to enhance the reliability of prediction. The prediction head was implemented by a Multiple Layer Perceptron with one hidden layer. In compliance with clinical studies, gender information was an additional input to the prediction head by embedded into the feature vector calculated from the backbone model. After the experimental comparison study, the best results showed a mean absolute error of 6.2 months on the public RSNA dataset and 5.1 months on the additional dataset using MobileNetV3 as the backbone.

Evaluation of Bone Age in Children: A Mini-Review

Bone age represents a common index utilized in pediatric radiology and endocrinology departments worldwide for the definition of skeletal maturity for medical and non-medical purpose. It is defined by the age expressed in years that corresponds to the level of maturation of bones. Although several bones have been studied to better define bone age, the hand and wrist X-rays are the most used images. In fact, the images obtained by hand and wrist X-ray reflect the maturity of different types of bones of the skeletal segment evaluated. This information, associated to the characterization of the shape and changes of bone components configuration, represent an important factor of the biological maturation process of a subject. Bone age may be affected by several factors, including gender, nutrition, as well as metabolic, genetic, and social factors and either acute and chronic pathologies especially hormone alteration. As well several differences can be characterized according to the numerous standardized methods developed over the past decades. Therefore, the complete characterization of the main methods and procedure available and particularly of all their advantages and disadvantages need to be known in order to properly utilized this information for all its medical and non-medical main fields of application.

Age estimation in the living: A scoping review of population data for skeletal and dental methods

Age estimation of living individuals has become a crucial part of the forensic practice, especially due to the global increase in cross-border migration. The low rate of birth registration in many countries, hence of identification documents of migrants, especially in Africa and Asia, highlights the importance of reliable methods for age estimation of living individuals. Despite the fact that a number of skeletal and dental methods for age estimation have been developed, their main limitation is that they are based on specific reference samples and there is still no consensus among researchers on whether these methods can be applied to all populations. Though this issue remains still unsolved, population information at a glance could be useful for forensic practitioners dealing with such issues. This study aims at presenting a scoping review and mapping of the current situation concerning population data for skeletal (hand-wrist and clavicle) and dental methods (teeth eruption and third molar formation) for age estimation in the living. Two hundred studies on the rate of skeletal maturation and four hundred thirty-nine on the rate of dental maturation were found, covering the period from 1952 and 2020 for a total of ninety-eight countries. For most of the western and central African countries there are currently no data on the rate of skeletal and dental maturation. The same applies to the countries of the Middle East, as well as the eastern European countries, especially as regard the skeletal development.

Are Automated and Visual Greulich and Pyle-Based Methods Applicable to Caucasian European Children With a Moroccan Ethnic Origin When Assessing Bone Age?

Introduction To test the accuracy of the visual and automated bone age assessment base on the Greulich and Pyle (GP) method in healthy Caucasian European children with a Moroccan ethnic origin. Material and methods Moroccan Caucasian (MC) children were retrospectively and consecutively enrolled along with age- and sex-matched control group (CG) of European Caucasian (EC) children enrolled from the general population. The two groups included 423 children aged from 2 to 15 years with a normal left-hand radiograph performed to rule out a trauma between March 2008 and December 2017. One radiologist, blinded to the BoneXpert^® (Visiana, Holte, Denmark) estimates, visually reviewed the radiographs using the GP atlas. The BoneXpert^® automatically analysed all 423 radiographs. The intraclass correlation coefficient (ICC), linear regression and Bland-Altman plots were performed to describe the agreement between each method and the chronological age (CA) and the agreement between the two methods. Results Visual bone age assessment was related to the CA in both girls (MC ICC 0.97; EC ICC 0.97) and boys (MC ICC 0.95; EC ICC 0.96). Automated bone age assessment was related to the CA in both girls (MC ICC 0.97; EC ICC 0.96) and boys (MC ICC 0.88; EC ICC 0.96). Bland-Altman plots showed an excellent agreement between the two methods in both sexes and ethnicities before puberty especially in Moroccan boys. Conclusion Visual and automatic bone age assessment based on the GP method, previously validated in the general population of Caucasian European children, can be confidently used in healthy Caucasian European children with a Moroccan ethnic origin.

The effect of vitamin D supplementation and nutritional intake on skeletal maturity and bone health in socio-economically deprived children

Purpose

1. To determine the effect of vitamin D supplementation on bone age (BA), a marker of skeletal maturity, and Bone Health Index (BHI), a surrogate marker of bone density. 2. To characterise the differences in nutritional intake and anthropometry between children with advanced vs. delayed BA.

Methods

The current study is a post hoc analysis of radiographs obtained as part of a randomised controlled trial. In this double-blind, placebo-controlled trial, deprived Afghan children (n = 3046) aged 1–11 months were randomised to receive six doses of oral placebo or vitamin D3 (100,000 IU) every 3 months for 18 months. Dietary intake was assessed through semi-quantitative food frequency questionnaires at two time points. Anthropometric measurements were undertaken at baseline and 18 months. Serum 25OHD was measured at five time points on a random subset of 632 children. Knee and wrist radiographs were obtained from a random subset (n = 641), of which 565 wrist radiographs were digitised for post-hoc analysis of BA and BHI using BoneXpert version 3.1.

Results

Nearly 93% (522, male = 291) of the images were analysable. The placebo (n = 258) and vitamin D (n = 264) groups were comparable at baseline. The mean (± SD) age of the cohort was 2 (± 0.3) years. At study completion, there was no difference in mean 25-hydroxy vitamin D concentrations [47 (95% CI 41, 56) vs. 55 (95% CI 45, 57) nmol/L, p = 0.2], mean (± SD) BA SDS [− 1.04 (1.36) vs. − 1.14 (1.26) years, p = 0.3] or mean (± SD) BHI SDS [− 0.30 (0.86) vs. − 0.31 (0.80), p = 0.8] between the placebo and vitamin D groups, respectively. Children with advanced skeletal maturity (BA SDS ≥ 0) when compared to children with delayed skeletal maturity (BA SDS < 0), had consumed more calories [mean (± SD) calories 805 (± 346) vs 723 (± 327) kcal/day, respectively, p < 0.05], were significantly less stunted (height SDS − 1.43 vs. − 2.32, p < 0.001) and underweight (weight SDS − 0.82 vs. − 1.45, p < 0.001), with greater growth velocity (11.57 vs 10.47 cm/ year, p < 0.05).

Conclusion

Deprived children have significant delay in skeletal maturation but no substantial impairment in bone health as assessed by BHI. BA delay was influenced by total calorie intake, but not bolus vitamin D supplementation.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00394-021-02511-5.

Skeletal Maturation in the Current Pediatric Mexican Population

OBJECTIVE

The most commonly used methods for bone age (BA) reading were described in the Caucasian population decades ago. However, there are secular trends in skeletal maturation and different BA patterns between ethnic groups. Automated BA reading makes updating references easier and more precise than human reading. The objective of the present study was to present automated BA reference curves according to chronological age and gender in the Mexican population and compare the maturation tempo with that of other populations.

METHODS

The study included 923 healthy participants aged 5 to 18 years between 2017 and 2018. A hand radio-graph was analyzed using BoneXpert software to obtain the automated BA reading according to Greulich and Pyle (G&P) and Tanner-Whitehouse 2 (TW2) references. We constructed reference curves using the average difference between the BA and chronological age according to sex and age.

RESULTS

The G&P and TW2 automated reference curves showed that Mexican boys exhibit delays in BA during middle childhood by 0.5 to 0.7 (95% confidence interval &lsqb;CI], -0.9 to -0.2) years; however, they demonstrate an advanced BA of up to 1.1 (95% CI, 0.8 to 1.4) years at the end of puberty. Mexican girls exhibited a delay in BA by 0.3 to 0.6 (95% CI, -0.9 to -0.1) years before puberty and an advanced BA of up to 0.9 (95% CI, 0.7 to 1.2) years at the end of puberty.

CONCLUSION

Mexican children aged <10 years exhibited a delay in skeletal maturity, followed by an advanced BA by approximately 1 year at the end of puberty. This may affect the estimation of growth potential in this population.

Magnetic resonance imaging for forensic age estimation in living children and young adults: a systematic review

The use of MRI in forensic age estimation has been explored extensively during the last decade. The authors of this paper synthesized the available MRI data for forensic age estimation in living children and young adults to provide a comprehensive overview that can guide age estimation practice and future research. To do so, the authors searched MEDLINE, Embase and Web of Science, along with cited and citing articles and study registers. Two authors independently selected articles, conducted data extraction, and assessed risk of bias. They considered study populations including living subjects up to 30 years old. Fifty-five studies were included in qualitative analysis and 33 in quantitative analysis. Most studies had biases including use of relatively small European (Caucasian) populations, varying MR approaches and varying staging techniques. Therefore, it was not appropriate to pool the age distribution data. The authors found that reproducibility of staging was remarkably lower in clavicles than in any other anatomical structure. Age estimation performance was in line with the gold standard, radiography, with mean absolute errors ranging from 0.85 years to 2.0 years. The proportion of correctly classified minors ranged from 65% to 91%. Multifactorial age estimation performed better than that based on a single anatomical site. The authors found that more multifactorial age estimation studies are necessary, together with studies testing whether the MRI data can safely be pooled. The current review results can guide future studies, help medical professionals to decide on the preferred approach for specific cases, and help judicial professionals to interpret the evidential value of age estimation results.

Evaluation of the clinical efficacy of a TW3-based fully automated bone age assessment system using deep neural networks

Purpose

The aim of this study was to evaluate the clinical efficacy of a Tanner-Whitehouse 3 (TW3)-based fully automated bone age assessment system on hand-wrist radiographs of Korean children and adolescents.

Materials and Methods

Hand-wrist radiographs of 80 subjects (40 boys and 40 girls, 7–15 years of age) were collected. The clinical efficacy was evaluated by comparing the bone ages that were determined using the system with those from the reference standard produced by 2 oral and maxillofacial radiologists. Comparisons were conducted using the paired t-test and simple regression analysis.

Results

The bone ages estimated with this bone age assessment system were not significantly different from those obtained with the reference standard (P>0.05) and satisfied the equivalence criterion of 0.6 years within the 95% confidence interval (− 0.07 to 0.22), demonstrating excellent performance of the system. Similarly, in the comparisons of gender subgroups, no significant difference in bone age between the values produced by the system and the reference standard was observed (P>0.05 for both boys and girls). The determination coefficients obtained via regression analysis were 0.962, 0.945, and 0.952 for boys, girls, and overall, respectively (P=0.000); hence, the radiologist-determined bone ages and the system-determined bone ages were strongly correlated.

Conclusion

This TW3-based system can be effectively used for bone age assessment based on hand-wrist radiographs of Korean children and adolescents.