RESUMEN
Background: Bone age assessment (BAA) is crucial for the diagnosis of growth disorders and the optimization of treatments. However, the random error caused by different observers' experiences and the low consistency of repeated assessments harms the quality of such assessments. Thus, automated assessment methods are needed. Methods: Previous research has sought to design localization modules in a strongly or weakly supervised fashion to aggregate part regions to better recognize subtle differences. Conversely, we sought to efficiently deliver information between multi-granularity regions for fine-grained feature learning and to directly model long-distance relationships for global understanding. The proposed method has been named the "Multi-Granularity and Multi-Attention Net (2M-Net)". Specifically, we first applied the jigsaw method to generate related tasks emphasizing regions with different granularities, and we then trained the model on these tasks using a hierarchical sharing mechanism. In effect, the training signals from the extra tasks created as an inductive bias, enabling 2M-Net to discover task relatedness without the need for annotations. Next, the self-attention mechanism acted as a plug-and-play module to effectively enhance the feature representation capabilities. Finally, multi-scale features were applied for prediction. Results: A public data set of 14,236 hand radiographs, provided by the Radiological Society of North America (RSNA), was used to develop and validate 2M-Net. In the public benchmark testing, the mean absolute error (MAE) between the bone age estimates of the model and of the reviewer was 3.98 months (3.89 months for males and 4.07 months for females). Conclusions: By using the jigsaw method to construct a multi-task learning strategy and inserting the self-attention module for efficient global modeling, we established 2M-Net, which is comparable to the previous best method in terms of performance.
RESUMEN
PURPOSE: Evaluate the agreement between bone age assessments conducted by two distinct machine learning system and standard Greulich and Pyle method. MATERIALS AND METHODS: Carpal radiographs of 225 patients (mean age 8 years and 10 months, SD = 3 years and 1 month) were retrospectively analysed at two separate institutions (October 2018 and May 2022) by both expert radiologists and radiologists in training as well as by two distinct AI software programmes, 16-bit AItm and BoneXpert® in a blinded manner. RESULTS: The bone age range estimated by the 16-bit AItm system in our sample varied between 1 year and 1 month and 15 years and 8 months (mean bone age 9 years and 5 months SD = 3 years and 3 months). BoneXpert® estimated bone age ranged between 8 months and 15 years and 7 months (mean bone age 8 years and 11 months SD = 3 years and 3 months). The average bone age estimated by the Greulich and Pyle method was between 11 months and 14 years, 9 months (mean bone age 8 years and 4 months SD = 3 years and 3 months). Radiologists' assessments using the Greulich and Pyle method were significantly correlated (Pearson's r > 0.80, p < 0.001). There was no statistical difference between BoneXpert® and 16-bit AItm (mean difference = - 0.19, 95%CI = (- 0.45; 0.08)), and the agreement between two measurements varies between - 3.45 (95%CI = (- 3.95; - 3.03) and 3.07 (95%CI - 3.03; 3.57). CONCLUSIONS: Both AI methods and GP provide correlated results, although the measurements made by AI were closer to each other compared to the GP method.
RESUMEN
PURPOSE: The aim of this study was to observe the influence of differential nutritional status on bone age (BA) change according to body mass index (BMI) and analyze the risk of advanced bone age in children with overweight and obesity. METHODS: In total 23,305 children from Beijing were included in this cross-sectional study. Childhood overweight and obesity were defined according to the China and World Health Organization growth criteria. The data were analyzed by the R coding platform version 4.3.0. RESULTS: Under the Chinese criteria, 29%, 15%, and 4% of boys with overweight; 33%, 33%, and 3% of boys with obesity; 39%, 25%, and 2% of girls with overweight; and 37%, 42% and 1% of girls with obesity had advanced, significantly advanced and delayed BA, respectively. After adjustment, overweight (odds ratio, 95% confidence interval, P under the Chinese criteria: 2.52, 2.30-2.75, <0.001 and 4.54, 4.06-5.09, <0.001) and obesity (4.31, 3.85-4.82, <0.001 and 14.01, 12.39-15.85, <0.001) were risk factors for both advanced BA and significantly advanced BA. CONCLUSIONS: Different nutritional statuses lead to differences in children's BA development. Children with overweight and obesity have higher rates of advanced BA under two growth criteria, and girls have more advances in BA than boys do. Overweight and obesity are risk factors for advanced BA.
RESUMEN
The cervical vertebral maturation (CVM) method is essential to determine the timing of orthodontic and orthopedic treatment. In this paper, a target detection model called DC-YOLOv5 is proposed to achieve fully automated detection and staging of CVM. A total of 1800 cephalometric radiographs were labeled and categorized based on the CVM stages. We introduced a model named DC-YOLOv5, optimized for the specific characteristics of CVM based on YOLOv5. This optimization includes replacing the original bounding box regression loss calculation method with Wise-IOU to address the issue of mutual interference between vertical and horizontal losses in Complete-IOU (CIOU), which made model convergence challenging. We incorporated the Res-dcn-head module structure to enhance the focus on small target features, improving the model's sensitivity to subtle sample differences. Additionally, we introduced the Convolutional Block Attention Module (CBAM) dual-channel attention mechanism to enhance focus and understanding of critical features, thereby enhancing the accuracy and efficiency of target detection. Loss functions, precision, recall, mean average precision (mAP), and F1 scores were used as the main algorithm evaluation metrics to assess the performance of these models. Furthermore, we attempted to analyze regions important for model predictions using gradient Class Activation Mapping (CAM) techniques. The final F1 scores of the DC-YOLOv5 model for CVM identification were 0.993, 0.994 for mAp0.5 and 0.943 for mAp0.5:0.95, with faster convergence, more accurate and more robust detection than the other four models. The DC-YOLOv5 algorithm shows high accuracy and robustness in CVM identification, which provides strong support for fast and accurate CVM identification and has a positive effect on the development of medical field and clinical diagnosis.
RESUMEN
PURPOSE: In pediatric medicine, precise estimation of bone age is essential for skeletal maturity evaluation, growth disorder diagnosis, and therapeutic intervention planning. Conventional techniques for determining bone age depend on radiologists' subjective judgments, which may lead to non-negligible differences in the estimated bone age. This study proposes a deep learning-based model utilizing a fully connected convolutional neural network(CNN) to predict bone age from left-hand radiographs. METHODS: The data set used in this study, consisting of 473 patients, was retrospectively retrieved from the PACS (Picture Achieving and Communication System) of a single institution. We developed a fully connected CNN consisting of four convolutional blocks, three fully connected layers, and a single neuron as output. The model was trained and validated on 80% of the data using the mean-squared error as a cost function to minimize the difference between the predicted and reference bone age values through the Adam optimization algorithm. Data augmentation was applied to the training and validation sets yielded in doubling the data samples. The performance of the trained model was evaluated on a test data set (20%) using various metrics including, the mean absolute error (MAE), median absolute error (MedAE), root-mean-squared error (RMSE), and mean absolute percentage error (MAPE). The code of the developed model for predicting the bone age in this study is available publicly on GitHub at https://github.com/afiosman/deep-learning-based-bone-age-estimation . RESULTS: Experimental results demonstrate the sound capabilities of our model in predicting the bone age on the left-hand radiographs as in the majority of the cases, the predicted bone ages and reference bone ages are nearly close to each other with a calculated MAE of 2.3 [1.9, 2.7; 0.95 confidence level] years, MedAE of 2.1 years, RMAE of 3.0 [1.5, 4.5; 0.95 confidence level] years, and MAPE of 0.29 (29%) on the test data set. CONCLUSION: These findings highlight the usability of estimating the bone age from left-hand radiographs, helping radiologists to verify their own results considering the margin of error on the model. The performance of our proposed model could be improved with additional refining and validation.
Asunto(s)
Determinación de la Edad por el Esqueleto , Aprendizaje Profundo , Humanos , Estudios Retrospectivos , Determinación de la Edad por el Esqueleto/métodos , Niño , Femenino , Masculino , Arabia Saudita , Adolescente , Preescolar , Lactante , Redes Neurales de la Computación , Huesos de la Mano/diagnóstico por imagen , Huesos de la Mano/crecimiento & desarrolloRESUMEN
Bone age determination in individuals is important for the diagnosis and treatment of growing children. This study aimed to develop a deep-learning model for bone age estimation using lateral cephalometric radiographs (LCRs) and regions of interest (ROIs) in growing children and evaluate its performance. This retrospective study included 1050 patients aged 4-18 years who underwent LCR and hand-wrist radiography on the same day at Pusan National University Dental Hospital and Ulsan University Hospital between January 2014 and June 2023. Two pretrained convolutional neural networks, InceptionResNet-v2 and NasNet-Large, were employed to develop a deep-learning model for bone age estimation. The LCRs and ROIs, which were designated as the cervical vertebrae areas, were labeled according to the patient's bone age. Bone age was collected from the same patient's hand-wrist radiograph. Deep-learning models trained with five-fold cross-validation were tested using internal and external validations. The LCR-trained model outperformed the ROI-trained models. In addition, visualization of each deep learning model using the gradient-weighted regression activation mapping technique revealed a difference in focus in bone age estimation. The findings of this comparative study are significant because they demonstrate the feasibility of bone age estimation via deep learning with craniofacial bones and dentition, in addition to the cervical vertebrae on the LCR of growing children.
Asunto(s)
Determinación de la Edad por el Esqueleto , Cefalometría , Vértebras Cervicales , Aprendizaje Profundo , Humanos , Niño , Determinación de la Edad por el Esqueleto/métodos , Vértebras Cervicales/diagnóstico por imagen , Vértebras Cervicales/anatomía & histología , Vértebras Cervicales/crecimiento & desarrollo , Cefalometría/métodos , Adolescente , Preescolar , Estudios Retrospectivos , Masculino , FemeninoRESUMEN
AIM AND OBJECTIVES: In forensic age estimation e.g. for judicial proceedings surpassed age thresholds can be legally relevant. To examine age related differences in skeletal development the recommendations by the Study Group on Forensic Age Diagnostics (AGFAD) are based on ionizing radiation (among others orthopantomograms, plain x-rays of the hand). Vieth et al. and Ottow et al. proposed MRI-classifications for the epiphyseal-diaphyseal fusion of the knee joint to define different age groups in healthy volunteers. The aim of the present study was to directly compare these two classifications in a large German patient population. MATERIALS AND METHODS: MRI of the knee joint of 900 patients (405 female, 495 male) from 10 to 28 years of age were retrospectively analyzed. Acquired T1-weighted turbo spin-echo sequence (TSE) and T2-weighted sequence with fat suppression by turbo inversion recovery magnitude (TIRM) were analyzed for the two classifications. The different bony fusion stages of the two classifications were determined and the corresponding chronological ages assigned. Differences between the sexes were analyzed. Intra- and inter-observer agreements were determined using Cohen's kappa. RESULTS: With the classification of Ottow et al. it was possible to determine completion of the 18th and 21st year of life in both sexes. With the classification of Vieth et al. completion of the 18th year of life for female patients and the 14th and 21st year of life in both sexes could be determined. The intra- and inter-observer agreement levels were very good (κ > 0.82). CONCLUSION: In the large German patient cohort of this study it was possible to determine the 18th year of life with for both sexes with the classification of Ottow et al. and for female patients with the classification of Vieth et al. It was also possible to determine the 21st year of life for all bones with the classification of Ottow et al. and for the distal femur with the classification of Vieth et al.
RESUMEN
BACKGROUND: Deviations between the determination of bone age (BA) according to Greulich and Pyle (G&P) and chronological age (CA) are common in Caucasians. Assessing these discrepancies in a population over time requires analysis of large samples and low intra-observer variability in BA estimation, both can be achieved with artificial intelligence-based software. The latest software-based reference curve contrasting the BA determined by G&P to the CA of Central European children dates back over two decades. OBJECTIVE: To examine whether the reference curve from a historical cohort from the Netherlands (Rotterdam cohort) between BA determined by G&P and CA still applies to a current Central European cohort and derive a current reference curve. MATERIALS AND METHODS: This retrospective single-center study included 1,653 children and adolescents (aged 3-17 years) who had received a radiograph of the hand following trauma. The G&P BA estimated using artificial intelligence-based software was contrasted with the CA, and the deviations were compared with the Rotterdam cohort. RESULTS: Among the participants, the mean absolute error between BA and CA was 0.92 years for girls and 0.97 years for boys. For the ages of 8 years (boys) and 11 years (girls) and upward, the mean deviation was significantly greater in the current cohort than in the Rotterdam cohort. The reference curves of both cohorts also differed significantly from each other (P < 0.001 for both boys and girls). CONCLUSION: The BA of the current Central European population and that of the curve from the Rotterdam cohort from over two decades ago differ. Whether this effect can be attributed to accelerated bone maturation needs further evaluation.
RESUMEN
Osteogenesis imperfecta (OI)is a rare genetically heterogeneous disorder caused by changes in the expression or processing of type I collagen. Clinical manifestations include bone fragility, decreased linear growth, and skeletal deformities that vary in severity. In typically growing children, skeletal maturation proceeds in a predictable pattern of changes in the size, shape, and mineralization on the hand and wrist bones that can be followed radiographically known at the bone age. Assessment of bone age can be clinically used to assess time remaining for linear growth, and the onset and duration of puberty, both of which can be useful in determining the timing of some surgeries or the interpretation of other imaging modalities such as bone densitometry. Additionally, deviations in the expected maturation process of the bone age may prompt or assist in the work up of a significant delay or advancement in a child's growth pattern. The primary aim of our study was to determine whether the bone age in children with a skeletal disorder such as OI follow the same pattern and rate of bone maturation compared to a control population. Using participants from the Natural History Study of the Brittle Bone Disorders Consortium, we analyzed 159 left hand and wrist radiographs (bone age) for a cross-sectional analysis and 55 bone ages repeated at approximately 24 months for a longitudinal analysis of skeletal maturation. Bone ages were read by a pediatric endocrinologist and by an automated analysis using a program called BoneXpert. Our results demonstrated that in children with mild-to-moderate OI (types I and IV), the skeletal maturation is comparable to chronological age-mated controls. For those with more severe forms of OI (type III), there is a delayed pattern of skeletal maturation of less than a year (10.5 months CI 5.1-16) P = 0.0012) at baseline and a delayed rate of maturation over the two-year follow up compared to type I (P = 0.06) and type III (P = 0.02). However, despite these parameters being statistically different, they may not be clinically significant. We conclude the bone age, with careful interpretation, can be used in the OI population in a way that is similar to the general pediatric population.
Asunto(s)
Desarrollo Óseo , Osteogénesis Imperfecta , Pubertad , Humanos , Osteogénesis Imperfecta/diagnóstico por imagen , Niño , Masculino , Femenino , Estudios Transversales , Estudios Longitudinales , Adolescente , Pubertad/fisiología , Determinación de la Edad por el Esqueleto , PreescolarRESUMEN
PURPOSE: Height age (HA) and bone age (BA) delay is well known in the patients with short stature. Therefore assessing pituitary hypoplasia based on chronological age (CA) might cause overdiagnosis of pituitary hypoplasia. We aimed to investigate the diagnostic and prognostic value of the PH and PV based on CA, HA, or BA in the patients with GHD. METHODS: Fifty-seven patients with severe and 40 patients with partial GHD and 39 patients with ISS assigned to the study. For defining the most accurate diagnosis of pituitary hypoplasia, PH and PV were evaluated based on CA, BA and HA. The relationship of each method with clinical features was examined. RESULTS: The mean PV was significantly larger in patients with ISS compared to the GH-deficient patients. PV was more correlated with clinical features including height SDS, stimulated GH concentration, IGF-1 and IGFBP-3 SDS, height velocity before and after rGH therapy. We found BA-based PV could discriminate GHD from ISS (Sensitivity: 17%, specificity: 98%, positive predictive value: 94%, negative predictive value: 39%), compared to the other methods based on PH or PV respect to CA and HA. 3% of patients with ISS, 17% of patients with GHD had pituitary hypoplasia based on PV-BA. CONCLUSION: PV based on BA, has the most accurate diagnostic value for defining pituitary hypoplasia. But it should be kept in mind that there might be still misdiagnosed patients by this method. PV is also a significant predictor for the rGH response.
RESUMEN
INTRODUCTION: HIV infection and its treatment compromises skeletal development (growth and maturation). Skeletal maturity is assessed as bone age (BA) on hand and wrist radiographs. BA younger than chronological age (CA) indicates delayed development. We conducted a cross-sectional study to determine differences between BA and CA (i.e., skeletal maturity deviation [SMD]), and risk factors associated with SMD in peripubertal children with and without HIV established on antiretroviral therapy (ART) including use of tenofovir disoproxil fumarate (TDF). METHODS: Children with HIV taking ART for at least two years and a comparison group of HIV-negative children, aged 8-16 years and frequency-matched by age and sex, were recruited from HIV clinics and local schools in the same catchment area, in Harare, Zimbabwe. BA was assessed from non-dominant hand-wrist radiographs using the Tanner Whitehouse 3 method. Negative SMD values correspond to delayed development, i.e., BA younger than CA. Multivariable linear regression models determined factors associated with SMD overall, and in children with HIV. RESULTS: In total, 534 participants (54% males) were included; by design CA was similar in males and females, whether living with or without HIV. Mean (SD) SMD was more negative in CWH than in HIV-negative children in both males [-1.4(1.4) vs. -0.4(1.1) years] and females [-1.1(1.3) vs. -0.0(1.2) years]. HIV infection and weight-for-age Z-score<-2 were associated with more negative SMD in both males and females after adjusting for socio-economic status, orphanhood, pubertal stage, and calcium intake. Age at ART initiation was associated with SMD in both males and females with those starting ART later more delayed: starting ART aged 4-8 years 1.14 (-1.84, -0.43), or over 8 years 1.47 (-2.30, -0.65) (p-value for trend < 0.001). Similar non-significant trends were seen in males. TDF exposure TDF exposure whether < 4years or ≥ 4 years was not associated with delayed development. CONCLUSION: Perinatally-acquired HIV infection and being underweight were independently associated with delayed skeletal maturation in both males and females. Starting ART later was independently associated with skeletal maturation delay in CWH. Given the known effects of delayed development on later health, it is important to find interventions to ensure healthy weight gain through early years and in CWH to initiate ART as early as possible.
Asunto(s)
Determinación de la Edad por el Esqueleto , Infecciones por VIH , Humanos , Estudios Transversales , Femenino , Masculino , Niño , Infecciones por VIH/tratamiento farmacológico , Zimbabwe/epidemiología , Adolescente , Desarrollo Óseo/efectos de los fármacos , Tenofovir/uso terapéutico , Factores de Riesgo , Fármacos Anti-VIH/uso terapéutico , Estudios de Casos y ControlesRESUMEN
BACKGROUND: Bone age assessment assists physicians in evaluating the growth and development of children. However, deep learning methods for bone age estimation do not currently incorporate differential features obtained through comparisons with other bone atlases. OBJECTIVE: To propose a more accurate method, Delta-Age-Sex-AdaIn (DASA-net), for bone age assessment, this paper combines age and sex distribution through adaptive instance normalization (AdaIN) and style transfer, simulating the process of visually comparing hand images with a standard bone atlas to determine bone age. MATERIALS AND METHODS: The proposed Delta-Age-Sex-AdaIn (DASA-net) consists of four modules: BoneEncoder, Binary code distribution, Delta-Age-Sex-AdaIn, and AgeDecoder. It is compared with state-of-the-art methods on both a public Radiological Society of North America (RSNA) pediatric bone age prediction dataset (14,236 hand radiographs, ranging from 1 to 228 months) and a private bone age prediction dataset from Zigong Fourth People's Hospital (474 hand radiographs, ranging from 12 to 218 months, 268 male). Ablation experiments were designed to demonstrate the necessity of incorporating age distribution and sex distribution. RESULTS: The DASA-net model achieved a lower mean absolute deviation (MAD) of 3.52 months on the RSNA dataset, outperforming other methods such as BoneXpert, Deeplasia, BoNet, and other deep learning based methods. On the private dataset, the DASA-net model obtained a MAD of 3.82 months, which is also superior to other methods. CONCLUSION: The proposed DASA-net model aided the model's learning of the distinctive characteristics of hand bones of various ages and both sexes by integrating age and sex distribution into style transfer.
RESUMEN
In Chinese criminal law, the ages of 12, 14, 16, and 18 years old play a significant role in the determination of criminal responsibility. In this study, we developed an epiphyseal grading system based on magnetic resonance image (MRI) of the hand and wrist for the Chinese Han population and explored the feasibility of employing deep learning techniques for bone age assessment based on MRI of the hand and wrist. This study selected 282 Chinese Han Chinese males aged 6.0-21.0 years old. In the course of our study, we proposed a novel deep learning model for extracting and enhancing MRI hand and wrist bone features to enhance the prediction of target MRI hand and wrist bone age and achieve precise classification of the target MRI and regression of bone age. The evaluation metric for the classification model including precision, specificity, sensitivity, and accuracy, while the evaluation metrics chosen for the regression model are MAE. The epiphyseal grading was used as a supervised method, which effectively solved the problem of unbalanced sample distribution, and the two experts showed strong consistency in the epiphyseal plate grading process. In the classification results, the accuracy in distinguishing between adults and minors was 91.1%, and the lowest accuracy in the three minor classifications (12, 14, and 16 years of age) was 94.6%, 91.1% and 96.4%, respectively. The MAE of the regression results was 1.24 years. In conclusion, the deep learning model proposed enabled the age assessment of hand and wrist bones based on MRI.
RESUMEN
INTRODUCTION: Automated bone age assessment (BAA) is of growing interest because of its accuracy and time efficiency in daily practice. In this study, we validated the clinical applicability of a commercially available artificial intelligence (AI)-powered X-ray bone age analyzer equipped with a deep learning-based automated BAA system and compared its performance with that of the Tanner-Whitehouse 3 (TW-3) method. METHODS: Radiographs prospectively collected from 30 centers across various regions in China, including 900 Chinese children and adolescents, were assessed independently by six doctors (three experts and three residents) and an AI analyzer for TW3 radius, ulna, and short bones (RUS) and TW3 carpal bone age. The experts' mean estimates were accepted as the gold standard. The performance of the AI analyzer was compared with that of each resident. RESULTS: For the estimation of TW3-RUS, the AI analyzer had a mean absolute error (MAE) of 0.48 ± 0.42. The percentage of patients with an absolute error of < 1.0 years was 86.78%. The MAE was significantly lower than that of rater 1 (0.54 ± 0.49, P = 0.0068); however, it was not significant for rater 2 (0.48 ± 0.48) or rater 3 (0.49 ± 0.46). For TW3 carpal, the AI analyzer had an MAE of 0.48 ± 0.65. The percentage of patients with an absolute error of < 1.0 years was 88.78%. The MAE was significantly lower than that of rater 2 (0.58 ± 0.67, P = 0.0018) and numerically lower for rater 1 (0.54 ± 0.64) and rater 3 (0.50 ± 0.53). These results were consistent for the subgroups according to sex, and differences between the age groups were observed. CONCLUSION: In this comprehensive validation study conducted in China, an AI-powered X-ray bone age analyzer showed accuracies that matched or exceeded those of doctor raters. This method may improve the efficiency of clinical routines by reducing reading time without compromising accuracy.
Assessing bone age, or how developed a child's skeleton is, is important in medical care, but the standard method can be time-consuming. Using AI to automatically assess bone age from X-ray images may improve efficiency without reducing accuracy. In this study, we evaluated how well an AI-powered X-ray bone age analyzer performed compared to the established TannerWhitehouse 3 (TW-3) method. X-ray images from 900 Chinese children and adolescents were collected from 30 centers. Six doctors (three experts, three residents) and the AI system independently assessed the TW-3 radius, ulna, and short bones (RUS) and TW-3 carpal bone age. The experts' assessments were considered the gold standard. The AI analyzer had an average error of 0.48 years for TW3-RUS bone age, with 87% of assessments within 1 year of the experts. For TW3 carpal bone age, the AI had an average error of 0.48 years, with 89% within 1 year. These results were similar to or better than those of the resident raters. These findings show the AI-powered analyzer can assess bone age as accurately as human raters. This technology may improve clinical efficiency by reducing the time required for bone age assessments without compromising accuracy.
RESUMEN
(1) Objective: In this study, a regression-based multi-modal deep learning model was developed for use in bone age assessment (BAA) utilizing hand radiographic images and clinical data, including patient gender and chronological age, as input data. (2) Methods: A dataset of hand radiographic images from 2974 pediatric patients was used to develop a regression-based multi-modal BAA model. This model integrates hand radiographs using EfficientNetV2S convolutional neural networks (CNNs) and clinical data (gender and chronological age) processed by a simple deep neural network (DNN). This approach enhances the model's robustness and diagnostic precision, addressing challenges related to imbalanced data distribution and limited sample sizes. (3) Results: The model exhibited good performance on BAA, with an overall mean absolute error (MAE) of 0.410, root mean square error (RMSE) of 0.637, and accuracy of 91.1%. Subgroup analysis revealed higher accuracy in females ≤ 11 years (MAE: 0.267, RMSE: 0.453, accuracy: 95.0%) and >11 years (MAE: 0.402, RMSE: 0.634, accuracy 92.4%) compared to males ≤ 13 years (MAE: 0.665, RMSE: 0.912, accuracy: 79.7%) and >13 years (MAE: 0.647, RMSE: 1.302, accuracy: 84.6%). (4) Conclusion: This model showed a generally good performance on BAA, showing a better performance in female pediatrics compared to male pediatrics and an especially robust performance in female pediatrics ≤ 11 years.
RESUMEN
Introduction: Bone age (BA) assessment is important in evaluating disorders of growth and puberty; the Greulich and Pyle atlas method (GP) is most used. We aimed to determine the weightage to be attributed by raters to various segments of the hand x-ray, namely, distal end of radius-ulna (RU), carpals, and short bones for rating bone age using the GP atlas method. Methods: 692 deidentified x-rays from a previous study (PUNE-dataset) and 400 from the Radiological Society of North America (RSNA-dataset) were included in the study. Mean of BA assessed by experienced raters was termed reference rating. Linear regression was used to model reference age as function of age ratings of the three segments. The root-mean-square-error (RMSE) of segmental arithmetic mean and weighted mean with respect to reference rating were computed for both datasets. Results: Short bones were assigned the highest weightage. Carpals were assigned higher weightage in pre-pubertal PUNE participants as compared to RSNA, vice-versa in RU segment of post-pubertal participants. The RMSE of weighted mean ratings was significantly lower than for the arithmetic mean in the PUNE dataset. Conclusion: We thus determined weightage to be attributed by raters to segments of the hand x-ray for assessment of bone age by the GP method.
RESUMEN
Skeletal radiographs along with dental examination are frequently used for age estimation in medicolegal cases where documentary evidence pertaining to age is not available. Wrist and hand radiographs are the most common skeletal radiograph considered for age estimation. Other parts imaged are elbow, shoulder, knee, and hip according to suspected age categories. Age estimation by wrist radiographs is usually done by the Tanner-Whitehouse method where the maturity level of each bone is categorized into stages and a final total score is calculated that is then transformed into the bone age. Careful assessment and interpretation at multiple joints are needed to minimize the error and categorize into age-group. In this article, we aimed to summarize a suitable radiographic examination and interpretation for bone age estimation in living children, adolescents, young adults, and adults for medicolegal purposes.
RESUMEN
This article will provide a perspective review of the most extensively investigated deep learning (DL) applications for musculoskeletal disease detection that have the best potential to translate into routine clinical practice over the next decade. Deep learning methods for detecting fractures, estimating pediatric bone age, calculating bone measurements such as lower extremity alignment and Cobb angle, and grading osteoarthritis on radiographs have been shown to have high diagnostic performance with many of these applications now commercially available for use in clinical practice. Many studies have also documented the feasibility of using DL methods for detecting joint pathology and characterizing bone tumors on magnetic resonance imaging (MRI). However, musculoskeletal disease detection on MRI is difficult as it requires multi-task, multi-class detection of complex abnormalities on multiple image slices with different tissue contrasts. The generalizability of DL methods for musculoskeletal disease detection on MRI is also challenging due to fluctuations in image quality caused by the wide variety of scanners and pulse sequences used in routine MRI protocols. The diagnostic performance of current DL methods for musculoskeletal disease detection must be further evaluated in well-designed prospective studies using large image datasets acquired at different institutions with different imaging parameters and imaging hardware before they can be fully implemented in clinical practice. Future studies must also investigate the true clinical benefits of current DL methods and determine whether they could enhance quality, reduce error rates, improve workflow, and decrease radiologist fatigue and burnout with all of this weighed against the costs.
Asunto(s)
Inteligencia Artificial , Enfermedades Musculoesqueléticas , Humanos , Enfermedades Musculoesqueléticas/diagnóstico por imagen , Imagen por Resonancia Magnética/métodos , Aprendizaje Profundo , Interpretación de Imagen Asistida por Computador/métodosRESUMEN
BACKGROUND: Serum IGF-1 is an important biochemical tool to diagnose and monitor GH-related disorders. However, ethnic-specific Indian data following consensus criteria for the establishment of normative data, are not available. Our objective was to generate chronological age (CA)-, bone age (BA)- and Tanner stage-specific normative data for IGF-1 in healthy Indian children and adolescents. METHODS: A cross-sectional epidemiological study was conducted in schools and the community, which enrolled apparently healthy children and adolescents with robust exclusion criteria. The outcome measure was serum IGF-1 assessed using an electro-chemiluminescence immunoassay (ECLIA). The 2.5th, 5th, 10th, 25th, 50th (median), 75th, 90th, 95th, and 97.5th centiles for IGF-1 were estimated using generalized additive models. RESULTS: We recruited 2226 apparently healthy participants and following exclusion, 1948 (1006 boys, 942 girls) were included in the final analysis. Girls had median IGF-1 peak at CA of 13 years (321.7â ng/mL), BA of 14 years (350.2â ng/mL) and Tanner stage IV (345â ng/mL), while boys had median IGF-1 peak at CA of 15 years (318.9â ng/mL) BA of 15 years (340.6â ng/mL) and Tanner stage III (304.8â ng/mL). Girls had earlier rise, peak and higher IGF-1 values. The reference interval (2.5th-97.5th percentile) was broader during peri-pubertal ages, indicating a higher physiological variability. CONCLUSION: This study provides ethnicity-specific normative data on serum IGF-1 and will improve the diagnostic utility of IGF-1 in the evaluation and management of growth disorders in Indian children and adolescents.
RESUMEN
In age determination, different methods aiming to obtain the closest result to chronological age have been investigated so far. The most commonly used one among these is the radiological method, which is usually used to evaluate the developmental stages of wrist bones or teeth. In our study, we assessed bone age estimations using the Gilsanz-Ratib atlas (GRA), which has recently become commonly used for children aged 9 to 15 years; evaluated the dental age, determined with Cameriere's European method; conducted morphometric measurements of the mandibular bone; and then examined their relationships with chronological age. The results of our study reveal that, in children during the puberty growth spurt, Cameriere's EU formula might have higher accuracy in estimating chronological age in younger age groups, while the GRA might be more accurate for older ages. Additionally, we conclude that of the mandibular morphometric measurements, condylar height and tangential ramus height show strong positive correlations with age. As a result, we conclude that the morphometric measurements evaluated in the present study can be used as auxiliary methods in forensic anthropology and forensic dentistry.