Potential Value of Conventional Ultrasound in Estimation of Bone Age in Patients from Birth to Near Adulthood

Bone age assessment based on three-dimensional ultrasound and artificial intelligence compared with paediatrician-read radiographic bone age: protocol for a prospective, diagnostic accuracy study

INTRODUCTION

Radiographic bone age (BA) assessment is widely used to evaluate children's growth disorders and predict their future height. Moreover, children are more sensitive and vulnerable to X-ray radiation exposure than adults. The purpose of this study is to develop a new, safer, radiation-free BA assessment method for children by using three-dimensional ultrasound (3D-US) and artificial intelligence (AI), and to test the diagnostic accuracy and reliability of this method.

METHODS AND ANALYSIS

This is a prospective, observational study. All participants will be recruited through Paediatric Growth and Development Clinic. All participants will receive left hand 3D-US and X-ray examination at the Shanghai Sixth People's Hospital on the same day, all images will be recorded. These image related data will be collected and randomly divided into training set (80% of all) and test set (20% of all). The training set will be used to establish a cascade network of 3D-US skeletal image segmentation and BA prediction model to achieve end-to-end prediction of image to BA. The test set will be used to evaluate the accuracy of AI BA model of 3D-US. We have developed a new ultrasonic scanning device, which can be proposed to automatic 3D-US scanning of hands. AI algorithms, such as convolutional neural network, will be used to identify and segment the skeletal structures in the hand 3D-US images. We will achieve automatic segmentation of hand skeletal 3D-US images, establish BA prediction model of 3D-US, and test the accuracy of the prediction model.

ETHICS AND DISSEMINATION

The Ethics Committee of Shanghai Sixth People's Hospital approved this study. The approval number is 2022-019. A written informed consent will be obtained from their parent or guardian of each participant. Final results will be published in peer-reviewed journals and presented at national and international conferences.

TRIAL REGISTRATION NUMBER

ChiCTR2200057236.

Si-MSPDNet: A multiscale Siamese network with parallel partial decoders for the 3-D measurement of spines in 3D ultrasonic images

Early screening and frequent monitoring effectively decrease the risk of severe scoliosis, but radiation exposure is a consequence of traditional radiograph examinations. Additionally, traditional X-ray images on the coronal or sagittal plane have difficulty providing three-dimensional (3-D) information on spinal deformities. The Scolioscan system provides an innovative 3-D spine imaging approach via ultrasonic scanning, and its feasibility has been demonstrated in numerous studies. In this paper, to further examine the potential of spinal ultrasonic data for describing 3-D spinal deformities, we propose a novel deep-learning tracker named Si-MSPDNet for extracting widely employed landmarks (spinous process (SP)) in ultrasonic images of spines and establish a 3-D spinal profile to measure 3-D spinal deformities. Si-MSPDNet has a Siamese architecture. First, we employ two efficient two-stage encoders to extract features from the uncropped ultrasonic image and the patch centered on the SP cut from the image. Then, a fusion block is designed to strengthen the communication between encoded features and further refine them from channel and spatial perspectives. The SP is a very small target in ultrasonic images, so its representation is weak in the highest-level feature maps. To overcome this, we ignore the highest-level feature maps and introduce parallel partial decoders to localize the SP. The correlation evaluation in the traditional Siamese network is also expanded to multiple scales to enhance cooperation. Furthermore, we propose a binary guided mask based on vertebral anatomical prior knowledge, which can further improve the performance of our tracker by highlighting the potential region with SP. The binary-guided mask is also utilized for fully automatic initialization in tracking. We collected spinal ultrasonic data and corresponding radiographs on the coronal and sagittal planes from 150 patients to evaluate the tracking precision of Si-MSPDNet and the performance of the generated 3-D spinal profile. Experimental results revealed that our tracker achieved a tracking success rate of 100% and a mean IoU of 0.882, outperforming some commonly used tracking and real-time detection models. Furthermore, a high correlation existed on both the coronal and sagittal planes between our projected spinal curve and that extracted from the spinal annotation in X-ray images. The correlation between the tracking results of the SP and their ground truths on other projected planes was also satisfactory. More importantly, the difference in mean curvatures was slight on all projected planes between tracking results and ground truths. Thus, this study effectively demonstrates the promising potential of our 3-D spinal profile extraction method for the 3-D measurement of spinal deformities using 3-D ultrasound data.

Bone Age Evaluated With Conventional Ultrasound: The Inter-Side Difference

OBJECTIVE

To confirm the inter-side difference of bone age evaluated with conventional ultrasound.

METHODS

In this prospective study, patients aged between 3 and 15 years with referral for bone age evaluation from the pediatric outpatient and inpatient department in our general medical institute were enrolled from March to October 2021. Bone ages from left and right side were evaluated with ultrasound using a summed ossification ratio of radius, ulna, and femur in all patients. For comparison, radiographic bone ages were assessed with X-ray films of the left hand and wrist.

RESULTS

A total of 192 patients were enrolled into the study, including 95 boys (10.5 years [3.2-14.6 years]) and 97 girls (8.7 years [3.7-15.4 years]). In the girls, the ossification ratio of the right radius was greater than the left (P = .030) and the ultrasonic bone age from the right side was more advanced than the left (P = .007). The differences of the left and right ultrasonic bone age were -0.66 to 0.61 years in the boys and -0.89 to 0.67 years in the girls. The differences of right ultrasonic bone age and radiographic bone age were -0.77 to 1.01 years in the boys and -1.12 to 1.14 years in the girls. The differences of left ultrasonic bone age and radiographic bone age were -0.73 to 0.91 years in the boys and -1.16 to 0.90 years in the girls.

CONCLUSIONS

Clinically important difference can be obtained from the left and right side when assessed bone age with the ossification ratio of bones by ultrasound, especially in girls. Both sides should be evaluated in clinical work.

Tanner-Whitehouse skeletal maturity score derived from ultrasound images to evaluate bone age

OBJECTIVE

The complexity of radiographic Tanner-Whitehouse method makes it less acceptable by radiologists and endocrinologists to assess bone age. Conventional ultrasound could be used to measure the ratio of the height of the ossification center to the epiphysis of the bone to evaluate maturity of bone. The purpose of this study is to obtain radiographic TW3 skeletal maturity score with ultrasound images.

METHODS

In this prospective diagnostic study, participants aged between 1 and 18 years undergoing radiography for bone age evaluation were evaluated from April 2019 to November 2021. Ultrasonic skeletal maturity scores of participants were transformed into radiographic skeletal maturity scores with the fitted formulas established in this study. Diagnostic performances of the transformed scores to diagnose advanced or delayed bone age were confirmed. Ultrasound images of 50 participants in the validation group were re-evaluated to confirm inter-rater reliability.

RESULTS

A total of 442 participants (median age, 9.5 years [interquartile range, 7.8-11.1 years]; 185 boys) were enrolled. Ultrasound determination of bone age had a sensitivity of 97% (34/35, 95% CI: 83, 99) and a specificity of 98% (106/108, 95% CI: 93, 99) to diagnose advanced or delayed bone age. The intra-class correlation coefficient for inter-rater reliability was 0.993 [95% CI: 0.988, 0.996], p < 0.0001.

CONCLUSIONS

Radiographic Tanner-Whitehouse skeletal maturity score could be obtained from ultrasound images in a simple, fast, accurate, and radiation-free manner.

KEY POINTS

• The fitting formulas between radiographic TW3 skeletal maturity score and ultrasonic skeletal maturity score were developed. • Through measurement of ossification ratios of bones with ultrasound, TW3 skeletal maturity score was obtained in a simple, fast, and radiation-free manner.

Ultrasound Imaging-Based Methods for Assessing Biological Maturity during Adolescence and Possible Application in Youth Sport: A Scoping Review

Bone maturity is an indicator for estimating the biological maturity of an individual. During adolescence, individuals show heterogeneous growth rates, and thus, differences in biological maturity should be considered in talent identification and development. Radiography of the left hand and wrist is considered the gold standard of biological maturity estimation. The use of ultrasound imaging (US) may be advantageous; however, its validity and reliability are under discussion. The aims of this scoping review are (1) to summarize the different methods for estimating biological maturity by US imaging in adolescents, (2) to obtain an overview of the level of validity and reliability of the methods, and (3) to point out the practicability and usefulness of ultrasound imaging in the field of youth sports. The search included articles published up to November 2022. The inclusion criteria stipulated that participants had to fall within the age range of 8 to 23 years and be free of bone disease and fractures in the region of interest. Nine body regions were investigated, while the hand and wrist were most commonly analyzed. US assessment methods were usually based on the estimation of a bone maturity stage, rather than a decimal bone age. Furthermore, 70% of the assessments were evaluated as applicable, 10% expressed restraint about implementation, and 20% were evaluated as not applicable. When tested, inter- and intra-rater reliability was high to excellent. Despite the absence of ionization, low costs, fast assessment, and accessibility, none of the US assessments could be referred to as a gold standard. If further development succeeds, its application has the potential to incorporate biological age into selection processes. This would allow for more equal opportunities in talent selection and thus make talent development fairer and more efficient.

Statistical Confirmation of a Method of US Determination of Bone Age

Background There is limited literature on conventional US to assess bone age. Purpose To determine the diagnostic performance of US in the assessment of abnormal bone age in Chinese children. Materials and Methods In this prospective study, children and young adults aged between birth and 19 years from a large provincial teaching hospital were enrolled from January to November 2020. Children without clinical diseases potentially affecting skeletal growth were included in the normal-value group. Children with clinically suspected growth disturbances who were undergoing bone age evaluation were included in the validation group. Ossification ratios (ie, the ratio of the height of the epiphyseal ossification center to the entire epiphysis, including the cartilaginous component) of the radius, ulna, and femur from all the children were measured using US. Ultrasonic skeletal maturity scores (ie, the summation of ossification ratios of the radius, ulna, and femur multiplied by 100) collected from children in the normal-value group were used for score-for-age curve fitting through Box-Cox power exponential distribution. Test performance characteristics for the ability of US to help diagnose abnormal bone age were determined using radiographic bone age as a reference standard. Statistically significant difference between groups was determined by using a paired-sample t test. Results A total of 1089 children (median age, 9 years [interquartile range, 3-14 years]; 578 boys) were enrolled, including 929 children (mean age, 8 years [interquartile range, 4-12 years]; 515 boys) in the normal-value group and 160 children (mean age, 9 years [interquartile range, 7-11 years]; 63 boys) in the validation group. Ultrasonic bone ages in the validation group were evaluated with use of the lists of normal score-for-age values. With radiographic bone age as a reference standard, US could help diagnose abnormal bone age with high sensitivity (93% [14 of 15 participants; 95% CI: 66, 100] for boys, 100% [14 of 14 participants; 95% CI: 73, 100] for girls) and specificity (98% [47 of 48 participants; 95% CI: 88, 100] for boys, 98% [81 of 83 participants; 95% CI: 91, 100] for girls). Conclusion The US scoring system established can be used to evaluate bone age with high sensitivity and specificity. Clinical trial registration no. ChiCTR1900027917 © RSNA, 2021 Online supplemental material is available for this article.

Ultrasonographic evaluation of the wrist and elbow joints: A pilot study to explore a non-invasive technique for age estimation

Age estimation is a vital aspect of the process of identification. Studying the appearance and fusion of long bones is one of the most commonly used methods for age estimation. Most research conducted on age estimation using the study of the appearance and fusion of ossification centres has been roentgenographic in nature. However, X-ray examination and computed tomography examination are associated with ionisation radiation. The present study investigated the use of ultrasonography (USG) as a means of visualising ossification centres of the elbow and wrist joints for age estimation in 31 Maharashtrian boys from Ahmednagar, India. The Schmeling et al. method of grading was used to score the fusion of ossification centres, and simple and multiple linear regression models were developed for age estimation. It was found that the ossification centres of the elbow and wrist joints followed a set pattern of maturation and fusion. The ossification centres of the elbow joint fused before the ossification centres of the wrist joint. The fusion scores of proximal radial epiphyses had the highest correlation with the decimal age of the participants, making its fusion the best indicator among all the ossification centres examined in this study for age estimation. Regression models to estimate age were generated using all the ossification centres. USG was found to be suitable for the purpose of age estimation based on ease of examination, minimal ionisation risks, its non-invasive nature and clear visualisation of ossification centres.

Summation of Ossification Ratios of Radius, Ulna and Femur: A New Parameter to Evaluate Bone Age by Ultrasound

Radiographic bone age (BA) assessment is reviewer dependent or time consuming. We aimed to clarify the correlation between sonographic ossification ratios (ORs, the ratio of ossification center and epiphysis diameters) of bones and radiographic BA and then to develop a repeatable parameter for BA assessment by ultrasound. The distal ends of the radius and ulna, medial epicondyle of the femur, medial tibial condyle, medial malleolus and lateral malleolus in 271 consecutive patients (132 boys and 139 girls) aged 0.1-19.0 y were imaged by ultrasound. The ORs of these bones were measured sonographically. The highest Pearson correlation r was that between the sum of the ORs of radius, ulna and femur (RUF) calculated from ultrasound images and the radiographic BA (0.97 in boys and 0.96 in girls). The entire process of collecting data and calculating the ORs of RUF took 2.6 ± 0.6 min. The ORs of RUF obtained with ultrasound have potential as an easy-to-perform and efficient quantitative assessment of BA.