中国组织工程研究 ›› 2025, Vol. 29 ›› Issue (35): 7511-7518.doi: 10.12307/2026.533

• 骨组织构建 bone tissue construction • 上一篇    下一篇

基于深度学习改进的膝骨关节炎自动诊断方法

方  颖1,张延伟2,利  晞3,颜培栋4,毕  苗1   

  1. 1广州中医药大学第三临床医学院,广东省广州市  510403;2广州中医药大学第三附属医院影像科,广东省广州市  510378;3广州医科大学附属第二医院放射科,广东省广州市  510260;4暨南大学附属珠海临床医学院,广东省珠海市  519099

  • 收稿日期:2024-10-29 接受日期:2024-12-31 出版日期:2025-12-18 发布日期:2025-04-30
  • 通讯作者: 张延伟,博士,主任医师,教授,硕士生导师,广州中医药大学第三附属医院影像科,广东省广州市 510378
  • 作者简介:方颖,女,1999年生,安徽省安庆市人,汉族,广州中医药大学在读硕士,主要从事肌骨影像学研究。

Improvements in automatic diagnosis methods for knee osteoarthritis based on deep learning

Fang Ying1, Zhang Yanwei2, Li Xi3, Yan Peidong4, Bi Miao1#br#   

  1. 1The Third Clinical Medical College of Guangzhou University of Chinese Medicine, Guangzhou 510403, Guangdong Province, China; 2Department of Imaging, the Third Affiliated Hospital, Guangzhou University of Chinese Medicine, Guangzhou 510378, Guangdong Province, China; 3Department of Radiology, Second Affiliated Hospital, Guangzhou University of Chinese Medicine, Guangzhou 510260, Guangdong Province, China; 4Zhuhai School of Clinical Medicine, Jinan University, Zhuhai 519009, Guangdong Province, China
  • Received:2024-10-29 Accepted:2024-12-31 Online:2025-12-18 Published:2025-04-30
  • Contact: Zhang Yanwei, PhD, Chief physician, Professor, Master’s supervisor, Department of Imaging, the Third Affiliated Hospital, Guangzhou University of Chinese Medicine, Guangzhou 510378, Guangdong Province, China
  • About author:Fang Ying, Master candidate, The Third Clinical Medical College of Guangzhou University of Chinese Medicine, Guangzhou 510403, Guangdong Province, China

摘要:


文题释义:
深度学习:是机器学习的一个分支,它通过构建和训练人工神经网络来学习和提取数据中的复杂模式和特征。与传统机器学习方法相比,深度学习能够处理大量的非结构化数据,如图像、音频和文本,从中自动提取特征,而无需依赖于人工特征工程。
膝骨关节炎:是一种常见的关节疾病,主要表现为膝关节的疼痛、僵硬和功能障碍。它通常是由于膝关节的软骨逐渐磨损、变薄,导致骨头之间的摩擦增加,从而引发炎症和疼痛。膝骨关节炎常见于中老年人,尤其是肥胖、关节受过伤或有遗传倾向的人群。

背景:膝骨关节炎是一种常见的退行性疾病,不仅严重影响患者的生活质量,同时增加社会医疗负担。早期准确诊断膝骨关节炎对于患者的治疗和预后至关重要,传统的诊断方法不仅主观且耗时,还不能保证稳定的高准确率。
目的:开发一种基于深度学习的膝骨关节炎自动诊断方法,利用深度学习网络提高诊断的准确性和效率。
方法:在YOLOv8n网络基础上采用 Efficient-ViT网络替换 YOLOv8n的骨干网络以及增加注意力机制的方法,提出了一种新的网络模型YOLOV8-ViT模型,用于自动识别和分类膝骨关节炎的X射线片图像。实验数据集来自广州中医药大学第三附属医院的5 078张膝骨关节炎患者的X射线片图像,由3个影像医师根据Kellgren-Lawrence分级标准采用labelme软件来标注膝关节炎部位并进行分类,采用并集结果。评价指标包括Precision、F1分数、mean average precision(mAP)、Recall、val/box_loss、val/cls_loss和val/dfl_loss。
结果与结论:实验结果表明,与YOLOv5n、YOLOv8n、YOLOv9n模型比较,YOLOV8-ViT模型的准确率、IoU阈值为0.5的平均精度(mAP50)、IoU阈值为0.5-0.95的平均精度(mAP50-95)、F1分数和Recall均有所提高,val/box_loss、val/cls_loss和val/dfl_loss分别降低了0.496、0.45和0.523,1.037、0.305和0.728,0.267、0654和0.854,验证了该模型具有较高的检测精度。

关键词: 膝骨关节炎, 深度学习, YOLOv8, Transformer, 目标检测, 检测精度

Abstract: BACKGROUND: Knee osteoarthritis is a common degenerative disease that significantly impacts patients' quality of life and increases the societal healthcare burden. Early and accurate diagnosis of knee osteoarthritis is crucial for the treatment and prognosis of patients. Traditional diagnostic methods are not only subjective and time-consuming but also do not guarantee consistently high accuracy.
OBJECTIVE: To develop an automatic diagnostic method for knee osteoarthritis based on deep learning, utilizing deep learning networks to improve diagnostic accuracy and efficiency.
METHODS: A new network model, YOLOV8-ViT, was proposed by replacing the backbone network of YOLOv8n with the Efficient-ViT network and incorporating attention mechanisms for the automatic identification and classification of X-ray images of knee osteoarthritis. The experimental dataset included 5 078 X-ray images of patients with knee osteoarthritis obtained from the Third Affiliated Hospital of Guangzhou University of Chinese Medicine. Three imaging physicians annotated the sites of knee osteoarthritis and classified them according to the Kellgren-Lawrence grading standard using Labelme software, and the results were combined. The evaluation indicators used in this study included Precision, F1 score, mean average precision (mAP), Recall, val/box_loss, val/cls_loss, and val/dfl_loss.
RESULTS AND CONCLUSION: The experimental results showed that the YOLOV8-ViT model outperformed the YOLOv5n, YOLOv8n, and YOLOv9n models in terms of precision, mAP50, mAP50-95, F1 score, and Recall, while lowering val/box_loss, val/cls_loss, and val/dfl_loss by 0.496, 0.45, and 0.523; 1.037, 0.305, and 0.728; and 0.267, 0.654, and 0.854, respectively. These experimental data validate that this model has high detection accuracy.

Key words: knee osteoarthritis, deep learning, YOLOv8, Transformer, object detection, detection precision

中图分类号: