Assessment of Bone Age Based on Hand Radiographs Using Regression-Based Multi-Modal Deep Learning
Jeoung Kun Kim1, Donghwi Park2, Min Cheol Chang3
1Department of Business Administration, School of Business, Yeungnam University, Gyeongsan-si 38541, Republic of Korea.
Life (Basel, Switzerland)
|June 27, 2024
Summary
A novel deep learning model accurately assesses bone age using hand X-rays and clinical data. The model demonstrated superior performance in females, particularly younger ones, highlighting its potential for pediatric bone age assessment.
Area of Science:
- Radiology
- Artificial Intelligence
- Pediatrics
Background:
- Accurate bone age assessment (BAA) is crucial for diagnosing and managing pediatric growth disorders.
- Traditional BAA methods can be subjective and time-consuming.
- Deep learning offers potential for objective and efficient BAA.
Purpose of the Study:
- To develop and evaluate a multi-modal deep learning model for bone age assessment.
- To integrate hand radiographic images with clinical data for improved BAA accuracy.
- To address data imbalance and sample size limitations in BAA.
Main Methods:
- A regression-based multi-modal deep learning model was developed using 2974 pediatric hand radiographs.
- Hand radiographs were processed using EfficientNetV2S convolutional neural networks (CNNs).
- Clinical data (gender, chronological age) were integrated via a deep neural network (DNN).
Main Results:
- The model achieved an overall mean absolute error (MAE) of 0.410 and accuracy of 91.1%.
- Higher accuracy was observed in females (≤11 years: MAE 0.267, 95.0% accuracy; >11 years: MAE 0.402, 92.4% accuracy).
- Males showed lower accuracy (≤13 years: MAE 0.665, 79.7% accuracy; >13 years: MAE 0.647, 84.6% accuracy).
Conclusions:
- The developed multi-modal deep learning model demonstrates good performance for bone age assessment.
- The model exhibits superior accuracy in female pediatric patients compared to males.
- The model shows particularly robust performance in female pediatrics aged 11 years and younger.
Related Concept Videos
01:40Bone Remodeling
34.4K
Bone remodeling is a continuous and balanced process of bone resorption by osteoclasts and bone formation by osteoblasts. In adults, it helps maintain bone mass and calcium homeostasis. While mechanical stress can stimulate turnover as part of the normal maintenance and reparative process, several hormones also regulate bone remodeling.
34.4K
01:18Classification of Bones
14.3K
The bones of the human skeletal system are of varied shapes, sizes, and functions. They can be classified based on their shape and function into four major classes: long bones, short bones, flat bones, and irregular bones. Some classifications include a fifth type, the sesamoid bones, as a separate class, whereas others categorize them under short bones.
Long and Short Bones
The appendicular skeleton, particularly the upper and lower limbs, is primarily made of long and short bones. The...
Long and Short Bones
The appendicular skeleton, particularly the upper and lower limbs, is primarily made of long and short bones. The...
14.3K
01:09Changes in the Appendicular Skeleton with Age
3.2K
The upper and lower limb initially develops as a small bulge called a limb bud, which appears on the lateral side of the early embryo. The upper limb bud appears near the end of the fourth week of development, with the lower limb bud appearing shortly after.
Initially, the limb buds consist of a core of mesenchyme covered by a layer of ectoderm. The ectoderm at the end of the limb bud thickens to form a narrow crest called the apical ectodermal ridge. This ridge stimulates the underlying...
Initially, the limb buds consist of a core of mesenchyme covered by a layer of ectoderm. The ectoderm at the end of the limb bud thickens to form a narrow crest called the apical ectodermal ridge. This ridge stimulates the underlying...
3.2K