• Korean Journal of Radiology
• /
• v.24 no.11
• /
• pp.1151-1163
• /
• 2023

Objective: To develop a deep-learning-based bone age prediction model optimized for Korean children and adolescents and evaluate its feasibility by comparing it with a Greulich-Pyle-based deep-learning model. Materials and Methods: A convolutional neural network was trained to predict age according to the bone development shown on a hand radiograph (bone age) using 21036 hand radiographs of Korean children and adolescents without known bone development-affecting diseases/conditions obtained between 1998 and 2019 (median age [interquartile range {IQR}], 9 [7-12] years; male:female, 11794:9242) and their chronological ages as labels (Korean model). We constructed 2 separate external datasets consisting of Korean children and adolescents with healthy bone development (Institution 1: n = 343; median age [IQR], 10 [4-15] years; male: female, 183:160; Institution 2: n = 321; median age [IQR], 9 [5-14] years; male: female, 164:157) to test the model performance. The mean absolute error (MAE), root mean square error (RMSE), and proportions of bone age predictions within 6, 12, 18, and 24 months of the reference age (chronological age) were compared between the Korean model and a commercial model (VUNO Med-BoneAge version 1.1; VUNO) trained with Greulich-Pyle-based age as the label (GP-based model). Results: Compared with the GP-based model, the Korean model showed a lower RMSE (11.2 vs. 13.8 months; P = 0.004) and MAE (8.2 vs. 10.5 months; P = 0.002), a higher proportion of bone age predictions within 18 months of chronological age (88.3% vs. 82.2%; P = 0.031) for Institution 1, and a lower MAE (9.5 vs. 11.0 months; P = 0.022) and higher proportion of bone age predictions within 6 months (44.5% vs. 36.4%; P = 0.044) for Institution 2. Conclusion: The Korean model trained using the chronological ages of Korean children and adolescents without known bone development-affecting diseases/conditions as labels performed better in bone age assessment than the GP-based model in the Korean pediatric population. Further validation is required to confirm its accuracy.

• Preventive Nutrition and Food Science
• /
• v.10 no.1
• /
• pp.40-45
• /
• 2005

Age-related changes in bone metabolism are well established by biochemical markers of bone matrix in serum and urine, but analysis of the residual bone matrix, which is still turning over, has not been investigated. In the present study, we measured in vivo rates of bone protein synthesis using a precursor-product method based on the exchange of ²H from ²H₂O into amino acids. Four percent ²H₂O was administered to mice in drinking water after intraperitonial (i.p) bolus injection of 99.9% ²H₂O. Mice were divided into the two groups: growing young mice were administered 4% ²H₂O for 12 weeks after an i.p bolus injection at 5 week of age, whereas weight stable adult mice started drinking 4% ²H₂O 8 weeks later than the growing group and continued 4% ²H₂O drinking for 8 weeks. Mass isotopomer abundance in alanine from bone protein was analyzed by gas chromatography/mass spectrometry. Body ²H₂O enrichments were in the range of 1.88-2.41% over the labeling period. The fractional synthesis rates (ks) of bone protein were 2.000±0.071%/d for growing mice and 0.243±0.014%/d for adult mice. These results demonstrate that the bone protein synthesis rate decreases with age and present direct evidence of age-related changes in bone protein synthesis.

• The Journal of Pediatrics of Korean Medicine
• /
• v.23 no.2
• /
• pp.145-157
• /
• 2009

Objectives : The purpose of this study was to examine relationships between bone age and body composition to make efficient clinical reviews on children's growth. Methods : 157 of children in age of 3 years to 16 years old were participated in this study(88 of boys and 69 of girls). They visited the department of pediatrics, OO university oriental hospital and were measured their body composition and bone age. Results : 1. An age and bone age, height, weight, and body mass index were positively correlated, and also a bone age and height, weight, and body mass index were positively correlated. 2. The level of soft lean mass, body fat mass, and MPH were increased in boys in higher height percentile. Children's predicted adult height was higher in children in higher height percentile. 3. The level of body fat mass was increased as weight percentile increased. Bone age, MPH was increased as weight percentile increased, especially in case of boys. In girl's case, the level of soft lean mass, their predicted adult height, the difference between children's bone age and their actual age was increased as weight percentile increased. Conclusions : Measuring bone age and body composition is the effective way to estimate children's growth and development in future.

• Journal of Korea Multimedia Society
• /
• v.12 no.3
• /
• pp.383-396
• /
• 2009

Bone age assessment has been widely used in pediatrics to identify endocrine problems of children. Since the number of trained doctors is far less than the demands, there has been numerous requests for automatic estimation of bone age. Therefore, in this paper, we propose an automatic bone age assessment method that utilizes pattern classification techniques. The proposed method consists of three modules; a finger segmentation module, a normalized shape model generation module and a bone age estimation module. The finger segmentation module segments fingers and epiphyseal regions by means of various image processing algorithms. The shape model abstraction module employ ASM to improves the accuracy of feature extraction for bone age estimation. In addition, SVM is used for estimation of bone age. Features for the estimation include the length of bone and the ratios of bone length. We evaluated the performance of the proposed method through statistical analysis by comparing the bone age assessment results by clinical experts and the proposed automatic method. Through the experimental results, the mean error of the assessment was 0.679 year, which was better than the average error acceptable in clinical practice.

• Korean Journal of Community Nutrition
• /
• v.3 no.3
• /
• pp.380-388
• /
• 1998

This study was conducted to investigate the distribution of bone density according to age and the prevalence rate of osteoporosis I 613, middle-aged women who visited Saint Bundo Hospital in Pusan from June to December, 1997. Mean bone density of lumbar spine(L2L4), and femoral neck of 50-59 years of age was significantly lower than those of 40-49 years of age(p<0.05). At the 60years of age, mean bone density of two sites were less than those of 50-59 years of age. Mean bone density of lumbar spine tin the group of sixties were 20.7% lower than that of group aged under 40 ; For femoral neck, women in their sixties showed 22.6% lower density compared to the women aged under forty. Bone density of ward's triangle of sixties were the least, which was 34.2% lower than that of group aged under 40. Bone density in lumbar spine, femoral neck, trochanter and ward's triangle correlates strongly with each other(p<0.001). The proportion of osteoporosis was 3.6% in the group of forties, 10.9% in the group of fifties and 33.8% for the group aged over 60, which was assessed by bone density of lumbar spine. Bone density of lumbar spine, femoral neck and ward's triangle were positively correlated with height, weight and BMI(p<0.001∼p<0.01), and weight showed highest correlation with the bone density. Forty-four percent of variation in lumbar spine bone density was explained by age and weight.

• The Journal of Korean Physical Therapy
• /
• v.15 no.4
• /
• pp.466-478
• /
• 2003

This study was to evaluate the body composition and bone mineral density according to aging in adult and investigated the relationship between various parameters such as body mass index(BMI), bone mineral density(BMD), bone mineral content(BMC), lean body mass(LBM), fat mass(FM) and the value obtained from dual energy X-ray absorptiometry(DEXA). The subjects were composed of healthy adult male and female who were $20^{\sim}73$ years old and they were divided three group according to age (A group : 20-39 yrs., B group : 40-59 yrs., C group : more than 60 yrs.). The conclusion derived from statistical analysis was as follows : 1. Bone mineral content and density were significantly affected by lean body mass(relatively, R=0.85 - 0.63). 2. There was significant difference among age groups in total bone mineral density. 3. There was significant difference among age groups in bone mineral content of male and female. 4. Lean body mass is diminished according to age, but there was not significant difference among age groups. 5. Fat mass of A group in male had the highest mass and followed by C group and B group. In female groups, fat mass of A group had the highest mass and followed by B group and C group. Abdominal fat mass is increased according to age. This result suggest that aging was closely relation with loss of muscle mass, bone mineral density and bone mineral content.

• Advances in nano research
• /
• v.12 no.2
• /
• pp.197-212
• /
• 2022

The purpose of this study is to develop an automatic software system for bone age evaluation and to evaluate its accuracy in testing and feasibility in clinical practice. 20394 left-hand radiographs of healthy children (2-18 years old) were collected from China Skeletal Development Survey data of 1998 and China Skeletal Development Survey data of 2005. Three experienced radiologists and China-05 standard maker jointly evaluate the stages of bone development and the reference bone age was determined by consensus. 1020 from 20394 radiographs were picked randomly as test set and the remaining 19374 radiographs as training set and validation set. Accuracy of the automatic software system for bone age assessment is evaluated in test set and two clinical test sets. Compared with the reference standard, the automatic software system based on RUS-CHN for bone age assessment has a 0.04 years old mean difference, ±0.40 years old in 95% confidence interval by single reading, a 85.6% percentage agreement of ratings, a 93.7% bone age accuracy rate, 0.17 years old of MAD, 0.29 years old of RMS; Compared with the reference standard, the automatic software system based on TW3-C RUS has a 0.04 years old mean difference, a ±0.38 years old in 95% confidence interval by single reading, a 90.9% percentage agreement of ratings, a 93.2% bone age accuracy rate, a 0.16 years of MAD, and a 0.28 years of RMS. Automatic software system, AI-China-05 showed reliably accuracy in bone age estimation and steady determination in different clinical test sets.

• Proceedings of the IEEK Conference
• /
• 2007.07a
• /
• pp.395-396
• /
• 2007

Bone age assessment has been widely used to measure the ossification in pediatric radiology. For the assessment, first bone age of each epiphyseal plate is estimated using DCT/LDA, then the bone age of a patient is calculated by using the median of 9 estimations. For some patients, however, due to various reasons such as X-ray image quality or the pose of fingers, it is common to miss couple of plates in automated systems. In this paper, we investigate the relationship between the number of detected plates and the accuracy of bone age assessment. In the experimental results, we confirmed the similarity between bone age assessed using more than 7 epiphyseal plates and that assessed using 9 epiphyseal plates.

• The Journal of Pediatrics of Korean Medicine
• /
• v.28 no.1
• /
• pp.43-51
• /
• 2014

Objectives The purpose of this study is to find out the relationship between mid parental height (MPH), birth weight, current growth condition of children (height, weight, BMI, body fat percentage, bone age) and final height of the future. Methods The study was conducted with 237 children, who were 12 - 14 years old. They were analyzed to find out the relationship between MPH, birth weight, height, current weight, BMI, body fat percentage, bone age and predicted height. Results 1. As MPH was increased, height and predicted height were also increased. As MPHs in girls were increased, 'bone age - chronological age' were decreased. As MPHs in girls were increased, body fat percentages were decreased. 2. As birth weights were increased, height, weight, BMI and body fat percentages were also increased in boys. 3. As body fat percentage was increased, predicted height was decreased. As 'bone age - chronological age' was increased, predicted height was decreased. As BMI was increased, 'bone age - chronological age' was increased. As body fat percentages in boys were increased, heights were decreased. As body fat percentages in girls were increased, 'bone age - chronological age' were increased. Conclusions MPH, birth weight, current growth condition (height, weight, BMI, body fat percentage, bone age) and predicted height are correlated to each other. There are some differences between boys and girls in these relationships.

• Asian-Australasian Journal of Animal Sciences
• /
• v.19 no.11
• /
• pp.1632-1637
• /
• 2006

We studied the changes in biochemical markers of bone metabolism in growing Thoroughbred horses. Serum osteocalcin (OC), as a marker for bone formation, and carboxy-terminal propeptide of type-I collagen (PICP), as a marker for bone formation, carboxy-terminal telopeptide of type-I collagen (ICTP), as a marker for bone resorption, were determined in nine clinically healthy horses from 3 d to 17 mo of age. The BW and withers height (WH) increased during the study. On the other hand, a rapid reduction in body weight gain (BWG) was observed between 1 mo and 9 mo of age and a rapid reduction in withers height gain was observed between 1 mo and 5 mo of age. The serum markers decreased significantly with increasing age. In particular, dramatic changes in serum markers occurred between 3 d to 1 wk and 5 to 7 mo of age in these horses, which suggests that bone turnover rapidly decreased after birth. On the other hand, the ratio of PICP to ICTP decreased through the experiment. This result suggests that the reduction in bone formation exceeded that of bone resorption. There was a significant correlation between markers and growth parameters, except for the correlation between PICP and BWG on single linear regression analysis. Serum OC and ICTP were affected by the WH in multiple linear regression analysis. These results indicated that the age-related variation in serum biochemical markers of bone metabolism reflected bone growth, but neither BW nor BWG. Therefore, we consider that changes in bone modeling are the major factor affecting the levels of serum biochemical markers by 17 mo of age in horses.