Bone age assessment based on deep convolution neural network incorporated with segmentation

Use of artificial intelligence in determination of bone age of the healthy individuals: A scoping review

BACKGROUND

Bone age assessment, as an indicator of biological age, is widely used in orthodontics and pediatric endocrinology. Owing to significant subject variations in the manual method of assessment, artificial intelligence (AI), machine learning (ML), and deep learning (DL) play a significant role in this aspect. A scoping review was conducted to search the existing literature on the role of AI, ML, and DL in skeletal age or bone age assessment in healthy individuals.

METHODS

A literature search was conducted in PubMed, Scopus, and Web of Science from January 2012 to December 2022 using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses-Extension for Scoping Reviews (PRISMA-ScR) and Joanna Briggs Institute guidelines. Grey literature was searched using Google Scholar and OpenGrey. Hand-searching of the articles in all the reputed orthodontic journals and the references of the included articles were also searched for relevant articles for the present scoping review.

RESULTS

Nineteen articles that fulfilled the inclusion criteria were included. Ten studies used skeletal maturity indicators based on hand and wrist radiographs, two studies used magnetic resonance imaging and seven studies used cervical vertebrae maturity indicators based on lateral cephalograms to assess the skeletal age of the individuals. Most of these studies were published in non-orthodontic medical journals. BoneXpert automated software was the most commonly used software, followed by DL models, and ML models in the studies for assessment of bone age. The automated method was found to be as reliable as the manual method for assessment.

CONCLUSIONS

This scoping review validated the use of AI, ML, or DL in bone age assessment of individuals. A more uniform distribution of sufficient samples in different stages of maturation, use of three-dimensional inputs such as magnetic resonance imaging, and cone beam computed tomography is required for better training of the models to generalize the outputs for use in the target population.

A deep learning approach for projection and body-side classification in musculoskeletal radiographs

BACKGROUND

The growing prevalence of musculoskeletal diseases increases radiologic workload, highlighting the need for optimized workflow management and automated metadata classification systems. We developed a large-scale, well-characterized dataset of musculoskeletal radiographs and trained deep learning neural networks to classify radiographic projection and body side.

METHODS

In this IRB-approved retrospective single-center study, a dataset of musculoskeletal radiographs from 2011 to 2019 was retrieved and manually labeled for one of 45 possible radiographic projections and the depicted body side. Two classification networks were trained for the respective tasks using the Xception architecture with a custom network top and pretrained weights. Performance was evaluated on a hold-out test sample, and gradient-weighted class activation mapping (Grad-CAM) heatmaps were computed to visualize the influential image regions for network predictions.

RESULTS

A total of 13,098 studies comprising 23,663 radiographs were included with a patient-level dataset split, resulting in 19,183 training, 2,145 validation, and 2,335 test images. Focusing on paired body regions, training for side detection included 16,319 radiographs (13,284 training, 1,443 validation, and 1,592 test images). The models achieved an overall accuracy of 0.975 for projection and 0.976 for body-side classification on the respective hold-out test sample. Errors were primarily observed in projections with seamless anatomical transitions or non-orthograde adjustment techniques.

CONCLUSIONS

The deep learning neural networks demonstrated excellent performance in classifying radiographic projection and body side across a wide range of musculoskeletal radiographs. These networks have the potential to serve as presorting algorithms, optimizing radiologic workflow and enhancing patient care.

RELEVANCE STATEMENT

The developed networks excel at classifying musculoskeletal radiographs, providing valuable tools for research data extraction, standardized image sorting, and minimizing misclassifications in artificial intelligence systems, ultimately enhancing radiology workflow efficiency and patient care.

KEY POINTS

• A large-scale, well-characterized dataset was developed, covering a broad spectrum of musculoskeletal radiographs. • Deep learning neural networks achieved high accuracy in classifying radiographic projection and body side. • Grad-CAM heatmaps provided insight into network decisions, contributing to their interpretability and trustworthiness. • The trained models can help optimize radiologic workflow and manage large amounts of data.

Forensic age estimation in living children: how accurate is the Greulich-Pyle method in Sabah, East Malaysia?

Background

The Greulich and Pyle's Radiographic Atlas of Skeletal Development of the Hand and Wrist (GP Atlas) is the most widely used method of determining the bone age (BA) of a child. It is also a widely accepted method for forensic age determination. As there is limited local bone age data for forensic age estimation, the purpose of this study was to assess the accuracy of the GP Atlas for forensic age determination in living Sabahan children.

Method

This study recruited 182 children between the ages of 9 years to 18 years. BA estimation of the left-hand anteroposterior radiographs were performed by two experienced radiologists using the Greulich-Pyle method.

Results

The BA estimates from two radiologists had very high interobserver reliability (ICC 0.937) and a strong positive interobserver correlation (r > 0.90). The GP method, significantly and consistently underestimated chronological age (CA) by 0.7, 0.6 and 0.7 years in overall children, boys and girls respectively with minimal errors. Mean absolute error and root of mean squared error for overall children was 1.5 and 2.2 years respectively, while mean absolute percentage error was 11.6%. This underestimation was consistent across all age groups but was statistically significant only at 13-13.9 and 17-18.9 years old age groups.

Conclusion

Despite high interobserver reliability of BA estimation using the GP Atlas, this method consistently underestimates the age of the child in all children to a significant degree, for both boys and girls across all age groups, with an acceptably low level of error metrics. Our findings suggest that locally validated GP Atlas or other type of assessments (artificial intelligence or machine learning) are needed for assessment of BA to accurately predict CA, since current GP Atlas standards significantly underestimated chronological age with minimal error for children in Sabah. A larger population-based study would be necessary for establishing a validated atlas of a bone age in Malaysia.

A Review Paper about Deep Learning for Medical Image Analysis

Medical imaging refers to the process of obtaining images of internal organs for therapeutic purposes such as discovering or studying diseases. The primary objective of medical image analysis is to improve the efficacy of clinical research and treatment options. Deep learning has revamped medical image analysis, yielding excellent results in image processing tasks such as registration, segmentation, feature extraction, and classification. The prime motivations for this are the availability of computational resources and the resurgence of deep convolutional neural networks. Deep learning techniques are good at observing hidden patterns in images and supporting clinicians in achieving diagnostic perfection. It has proven to be the most effective method for organ segmentation, cancer detection, disease categorization, and computer-assisted diagnosis. Many deep learning approaches have been published to analyze medical images for various diagnostic purposes. In this paper, we review the work exploiting current state-of-the-art deep learning approaches in medical image processing. We begin the survey by providing a synopsis of research works in medical imaging based on convolutional neural networks. Second, we discuss popular pretrained models and general adversarial networks that aid in improving convolutional networks' performance. Finally, to ease direct evaluation, we compile the performance metrics of deep learning models focusing on COVID-19 detection and child bone age prediction.

A novel approach to radiographic detection of growth development period with hand-wrist radiographs: A preliminary study with ImageJ imaging software

OBJECTIVE

The purpose of this study is to determine whether or not the ImageJ program can be used to automatically determine the growth period of the hand and wrist which have different growth-development periods according to the density values in the phalanges in radiographs.

SETTING AND SAMPLE POPULATION

Our study included hands-wrist radiographs of 270 individuals aged 8-17 years.

MATERIAL AND METHODS

The study's participants were classified into 7 groups according to their skeletal maturation stage (PP2=, MP3=, MP3cap, DP3u, PP3u, MP3u, and Ru) which included pre-peak, peak, and post-peak periods. The total density values (TDV) and pure density values (PDV) in the distal, medial, and proximal phalanges were calculated using each radiograph in the ImageJ program. Analysis of variance (ANOVA) was used to evaluate the density values and chronological age, and pairwise comparisons were made using the post-hoc LSD test.

RESULTS

The total density value was graphically zigzagged in the mesial, distal, and proximal phalanges. However, the pure density value increased continuously until the post-peak period and decreased after the DP3u period until the Ru period. While no significant difference in total density values was observed between the growth periods for all three phalanges, a significant difference in pure density values was observed.

CONCLUSION

It has been demonstrated in the ImageJ program that the peak growth period can be distinguished using the pure density values obtained from all phalanges of the third finger and that this method can be used as an alternative to the growth period detection through artificial intelligence.

Deep Learning-Based Medical Images Segmentation of Musculoskeletal Anatomical Structures: A Survey of Bottlenecks and Strategies

By leveraging the recent development of artificial intelligence algorithms, several medical sectors have benefited from using automatic segmentation tools from bioimaging to segment anatomical structures. Segmentation of the musculoskeletal system is key for studying alterations in anatomical tissue and supporting medical interventions. The clinical use of such tools requires an understanding of the proper method for interpreting data and evaluating their performance. The current systematic review aims to present the common bottlenecks for musculoskeletal structures analysis (e.g., small sample size, data inhomogeneity) and the related strategies utilized by different authors. A search was performed using the PUBMED database with the following keywords: deep learning, musculoskeletal system, segmentation. A total of 140 articles published up until February 2022 were obtained and analyzed according to the PRISMA framework in terms of anatomical structures, bioimaging techniques, pre/post-processing operations, training/validation/testing subset creation, network architecture, loss functions, performance indicators and so on. Several common trends emerged from this survey; however, the different methods need to be compared and discussed based on each specific case study (anatomical region, medical imaging acquisition setting, study population, etc.). These findings can be used to guide clinicians (as end users) to better understand the potential benefits and limitations of these tools.

Emerging Applications of Deep Learning in Bone Tumors: Current Advances and Challenges

Deep learning is a subfield of state-of-the-art artificial intelligence (AI) technology, and multiple deep learning-based AI models have been applied to musculoskeletal diseases. Deep learning has shown the capability to assist clinical diagnosis and prognosis prediction in a spectrum of musculoskeletal disorders, including fracture detection, cartilage and spinal lesions identification, and osteoarthritis severity assessment. Meanwhile, deep learning has also been extensively explored in diverse tumors such as prostate, breast, and lung cancers. Recently, the application of deep learning emerges in bone tumors. A growing number of deep learning models have demonstrated good performance in detection, segmentation, classification, volume calculation, grading, and assessment of tumor necrosis rate in primary and metastatic bone tumors based on both radiological (such as X-ray, CT, MRI, SPECT) and pathological images, implicating a potential for diagnosis assistance and prognosis prediction of deep learning in bone tumors. In this review, we first summarized the workflows of deep learning methods in medical images and the current applications of deep learning-based AI for diagnosis and prognosis prediction in bone tumors. Moreover, the current challenges in the implementation of the deep learning method and future perspectives in this field were extensively discussed.

A Study to Evaluate Accuracy and Validity of the EFAI Computer-Aided Bone Age Diagnosis System Compared With Qualified Physicians

Study Objectives

In previous research, we built a deep neural network model based on Inception-Resnet-v2 to predict bone age (EFAI-BAA). The primary objective of the study was to determine if the EFAI-BAA was substantially concordant with the qualified physicians in assessing bone ages. The secondary objective of the study was to determine if the EFAI-BAA was no different in the clinical rating (advanced, normal, or delayed) with the qualified physicians.

Method

This was a retrospective study. The left-hand X-ray images of male subjects aged 3-16 years old and female subjects aged 2-15 years old were collected from China Medical University Hospital (CMUH) and Asia University Hospital (AUH) retrospectively since the trial began until the included image amount reached 368. This was a blinded study. The qualified physicians who ran, read, and interpreted the tests were blinded to the values assessed by the other qualified physicians and the EFAI-BAA.

Results

The concordance correlation coefficient (CCC) between the EFAI-BAA (EFAI-BAA), the evaluation of bone age by physician in Kaohsiung Veterans General Hospital (KVGH), Taichung Veterans General Hospital (TVGH2), and in Taipei Tzu Chi Hospital (TZUCHI-TP) was 0.9828 (95% CI: 0.9790-0.9859, p-value = 0.6782), 0.9739 (95% CI: 0.9681-0.9786, p-value = 0.0202), and 0.9592 (95% CI: 0.9501-0.9666, p-value = 0.4855), respectively.

Conclusion

There was a consistency of bone age assessment between the EFAI-BAA and each one of the three qualified physicians (CCC = 0.9). As the significant difference in the clinical rating was only found between the EFAI-BAA and the qualified physician in TVGH2, the performance of the EFAI-BAA was considered similar to the qualified physicians.