Lightweight deep learning system for automated bone age assessment in Chinese children: enhancing clinical efficiency and diagnostic accuracy

Bone age assessment (BAA) is a critical diagnostic tool for evaluating skeletal maturity and monitoring growth disorders. Traditional clinical methods, however, are highly subjective, time-consuming, and reliant on clinician expertise, leading to inefficiencies and variability in accuracy. To addres...

Full description

Saved in:
Bibliographic Details
Main Authors: Pang Hai, Zhang Bin, Liu Kesheng, Li Cong, Xu Fei
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-07-01
Series:Frontiers in Endocrinology
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fendo.2025.1604133/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849719416663572480
author Pang Hai
Zhang Bin
Liu Kesheng
Li Cong
Xu Fei
author_facet Pang Hai
Zhang Bin
Liu Kesheng
Li Cong
Xu Fei
author_sort Pang Hai
collection DOAJ
description Bone age assessment (BAA) is a critical diagnostic tool for evaluating skeletal maturity and monitoring growth disorders. Traditional clinical methods, however, are highly subjective, time-consuming, and reliant on clinician expertise, leading to inefficiencies and variability in accuracy. To address these limitations, this study introduces a novel lightweight two-stage deep learning framework based on the Chinese 05 BAA standard. In the first stage, the YOLOv8 algorithm precisely localizes 13 key epiphyses in hand radiographs, achieving a mean Average Precision (mAP) of 99.5% at Intersection over Union (IoU) = 0.5 and 94.0% within IoU 0.5–0.95, demonstrating robust detection performance. The second stage employs a modified EfficientNetB3 architecture for fine-grained epiphyseal grade classification, enhanced by the Rectified Adam (RAdam) optimizer and a composite loss function combining center loss and weighted cross-entropy to mitigate class imbalance. The model attains an average accuracy of 80.3% on the training set and 81.5% on the test set, with a total parameter count of 15.8 million—56–86% fewer than comparable models (e.g., ResNet50, InceptionV3). This lightweight design reduces computational complexity, enabling faster inference while maintaining diagnostic precision. This framework holds transformative potential for pediatric endocrinology and orthopedics by standardizing BAA, improving diagnostic equity, and optimizing resource use. Success hinges on addressing technical, ethical, and adoption challenges through collaborative efforts among developers, clinicians, and regulators. Future directions might include multimodal AI integrating clinical data (e.g., height, genetics) for holistic growth assessments.
format Article
id doaj-art-5711c4ce78004c5dae7e3e1f5d3cd776
institution DOAJ
issn 1664-2392
language English
publishDate 2025-07-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Endocrinology
spelling doaj-art-5711c4ce78004c5dae7e3e1f5d3cd7762025-08-20T03:12:09ZengFrontiers Media S.A.Frontiers in Endocrinology1664-23922025-07-011610.3389/fendo.2025.16041331604133Lightweight deep learning system for automated bone age assessment in Chinese children: enhancing clinical efficiency and diagnostic accuracyPang HaiZhang BinLiu KeshengLi CongXu FeiBone age assessment (BAA) is a critical diagnostic tool for evaluating skeletal maturity and monitoring growth disorders. Traditional clinical methods, however, are highly subjective, time-consuming, and reliant on clinician expertise, leading to inefficiencies and variability in accuracy. To address these limitations, this study introduces a novel lightweight two-stage deep learning framework based on the Chinese 05 BAA standard. In the first stage, the YOLOv8 algorithm precisely localizes 13 key epiphyses in hand radiographs, achieving a mean Average Precision (mAP) of 99.5% at Intersection over Union (IoU) = 0.5 and 94.0% within IoU 0.5–0.95, demonstrating robust detection performance. The second stage employs a modified EfficientNetB3 architecture for fine-grained epiphyseal grade classification, enhanced by the Rectified Adam (RAdam) optimizer and a composite loss function combining center loss and weighted cross-entropy to mitigate class imbalance. The model attains an average accuracy of 80.3% on the training set and 81.5% on the test set, with a total parameter count of 15.8 million—56–86% fewer than comparable models (e.g., ResNet50, InceptionV3). This lightweight design reduces computational complexity, enabling faster inference while maintaining diagnostic precision. This framework holds transformative potential for pediatric endocrinology and orthopedics by standardizing BAA, improving diagnostic equity, and optimizing resource use. Success hinges on addressing technical, ethical, and adoption challenges through collaborative efforts among developers, clinicians, and regulators. Future directions might include multimodal AI integrating clinical data (e.g., height, genetics) for holistic growth assessments.https://www.frontiersin.org/articles/10.3389/fendo.2025.1604133/fullCH05bone age assessmentlightweight deep neural networkYOLOv8EfficientNetB3
spellingShingle Pang Hai
Zhang Bin
Liu Kesheng
Li Cong
Xu Fei
Lightweight deep learning system for automated bone age assessment in Chinese children: enhancing clinical efficiency and diagnostic accuracy
Frontiers in Endocrinology
CH05
bone age assessment
lightweight deep neural network
YOLOv8
EfficientNetB3
title Lightweight deep learning system for automated bone age assessment in Chinese children: enhancing clinical efficiency and diagnostic accuracy
title_full Lightweight deep learning system for automated bone age assessment in Chinese children: enhancing clinical efficiency and diagnostic accuracy
title_fullStr Lightweight deep learning system for automated bone age assessment in Chinese children: enhancing clinical efficiency and diagnostic accuracy
title_full_unstemmed Lightweight deep learning system for automated bone age assessment in Chinese children: enhancing clinical efficiency and diagnostic accuracy
title_short Lightweight deep learning system for automated bone age assessment in Chinese children: enhancing clinical efficiency and diagnostic accuracy
title_sort lightweight deep learning system for automated bone age assessment in chinese children enhancing clinical efficiency and diagnostic accuracy
topic CH05
bone age assessment
lightweight deep neural network
YOLOv8
EfficientNetB3
url https://www.frontiersin.org/articles/10.3389/fendo.2025.1604133/full
work_keys_str_mv AT panghai lightweightdeeplearningsystemforautomatedboneageassessmentinchinesechildrenenhancingclinicalefficiencyanddiagnosticaccuracy
AT zhangbin lightweightdeeplearningsystemforautomatedboneageassessmentinchinesechildrenenhancingclinicalefficiencyanddiagnosticaccuracy
AT liukesheng lightweightdeeplearningsystemforautomatedboneageassessmentinchinesechildrenenhancingclinicalefficiencyanddiagnosticaccuracy
AT licong lightweightdeeplearningsystemforautomatedboneageassessmentinchinesechildrenenhancingclinicalefficiencyanddiagnosticaccuracy
AT xufei lightweightdeeplearningsystemforautomatedboneageassessmentinchinesechildrenenhancingclinicalefficiencyanddiagnosticaccuracy