使用深度学习和对比转换的准确儿科骨年龄预测模型.
Dong Hyeok Choi1,2,3, So Hyun Ahn4,5, Rena Lee6
1Department of Medicine, Yonsei University College of Medicine, Seoul, Korea.
Ewha medical journal
|July 24, 2025
概括
深度学习模型使用手部X射线准确预测儿科骨年龄. 像图表平衡 (HE) 这样的图像预处理技术没有显著影响预测准确性,改善了临床生长评估.
科学领域:
- 放射学 放射学是一门学科.
- 人工智能的人工智能
- 儿科内分泌学 儿科内分泌学
背景情况:
- 准确的骨龄评估对于儿科生长监测和临床决策至关重要.
- 传统的骨年龄测定方法可能是主观的,耗时的.
研究的目的:
- 使用深度学习和对比转换技术开发一个准确的儿科骨年龄预测模型.
- 加强儿科生长评估中的临床决策.
主要方法:
- 利用各种深度学习模型 (CNN,ResNet50,VGG19,Inception V3,Xception) 在儿科手部X射线图像上进行训练.
- 应用对比转换技术 (模糊对比增强,CLAHE,HE) 用于图像预处理.
- 使用平均绝对误差 (MAE) 和根平均平方误差 (RMSE) 评估模型性能.
主要成果:
- 该Xception型号实现了最佳性能,MAE为41.12.
- 立体图平衡 (HE) 提高了图像质量,增强了SNR和对比度.
- 骨年龄预测的准确性得到改善,MAE从2.11降至0.24,RMSE从0.21降至0.02在预处理后.
结论:
- 深度学习模型,特别是Xception,显示了准确的儿科骨年龄预测的前景.
- 像HE这样的图像预处理技术可以提高图像质量,但不会显著改变预测性能.
- 这些发现支持整合人工智能驱动的工具,以改善儿科生长评估.
相关概念视频
01:18Classification of Bones
7.2K
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...
7.2K
01:40Bone Remodeling
38.5K
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.
38.5K