[1]蒋杰伟,丁科,冯洋阳,等.基于裂隙灯图像区域图像块的角膜炎自动诊断方法[J].中国医学物理学杂志,2025,42(9):1229-1235.[doi:DOI:10.3969/j.issn.1005-202X.2025.09.015]
　JIANG Jiewei,DING Ke,FENG Yangyang,et al.Automatic diagnosis method for keratitis based on regional image patches from slit-lamp images[J].Chinese Journal of Medical Physics,2025,42(9):1229-1235.[doi:DOI:10.3969/j.issn.1005-202X.2025.09.015]
点击复制

基于裂隙灯图像区域图像块的角膜炎自动诊断方法()
分享到：

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
42
期数:
2025年第9期
页码:
1229-1235
栏目:
医学人工智能
出版日期:
2025-09-30

文章信息/Info

Title:
Automatic diagnosis method for keratitis based on regional image patches from slit-lamp images
文章编号:
1005-202X（2025）09-1229-07
作者:
蒋杰伟1; 丁科1; 冯洋阳1; 辛宇1; 巩稼民1; 李中文2
1.西安邮电大学电子工程学院， 陕西 西安 710121； 2.温州医科大学附属宁波市眼科医院， 浙江 宁波 315000
Author(s):
JIANG Jiewei1;  DING Ke1;  FENG Yangyang1;  XIN Yu1;  GONG Jiamin1;  LI Zhongwen2
1. School of Electronic Engineering, Xian University of Posts and Telecommunications, Xian 710121, China 2. Ningbo Eye Hospital, Wenzhou Medical University, Ningbo 315000, China
关键词:
角膜炎; 自动诊断; 裂隙灯图像块; 特征融合; 卷积神经网络; 代价敏感
Keywords:
keratitis automatic diagnosis slit-lamp image patch feature fusion convolutional neural network cost-sensitive
分类号:
R318；TP391.7
DOI:
DOI:10.3969/j.issn.1005-202X.2025.09.015
文献标志码:
A
摘要:
为解决角膜炎人工诊断存在费时、费力、主观性强等缺点以及基于裂隙灯原始图像自动诊断准确率普遍偏低的问题，提出一种融合角膜病灶与结膜充血样并发症图像块特征的自动诊断方法，即从角膜区和结膜区采样，利用基于代价敏感的卷积神经网络提取并级联高层特征，借助主成分分析方法降维后输入全连接层分类。经在宁波市眼科医院收集的6 414张裂隙灯图像上训练评估，该方法在角膜炎、正常角膜、其他异常角膜上的准确率分别达97.8%、98.6%、97.0%，有效融合相关特征，为高准确率角膜炎诊断提供可行方案。
Abstract:
Abstract: A method that integrates the features of image patches from corneal lesions and conjunctival congestion-like complications is proposed to address the limitations of manual keratitis diagnosis (i.e., time-consuming, laborious, high subjectivity) and the generally low accuracy of automatic keratitis diagnosis based on original slit-lamp images. Specifically, samples are acquired from the corneal and conjunctival regions. A cost-sensitive convolutional neural network is then used to extract and concatenate the high-level features of these image patches. After dimensionality reduction through principal component analysis, the processed features are input into the fully connected layer for classification. Trained and evaluated on 6 414 slit-lamp images collected from Ningbo Eye Hospital, the proposed method achieves accuracies of 97.8%, 98.6%, and 97.0% for keratitis, normal cornea, and other abnormal corneas, respectively. This method effectively integrates relevant features and provides a feasible solution for high-accuracy keratitis diagnosis.

相似文献/References:

[1]翁羽洁,李忠贤,姬宇程,等.基于改进阈值的VGG网络的新冠肺炎CT图像自动诊断算法[J].中国医学物理学杂志,2022,39(6):731.[doi:DOI:10.3969/j.issn.1005-202X.2022.06.013]
　WENG Yujie,LI Zhongxian,JI Yucheng,et al.Automatic diagnosis algorithm for COVID-19 CT images using improved threshold-based VGG network[J].Chinese Journal of Medical Physics,2022,39(9):731.[doi:DOI:10.3969/j.issn.1005-202X.2022.06.013]

备注/Memo

备注/Memo:
【收稿日期】2025-03-10 【基金项目】国家自然科学基金（62276210，82201148，62376215）；陕西省重点研发计划项目（2025CY-YBXM-044）；陕西省教育厅科学研究计划项目（24JK0651）；浙江省自然科学基金（LQ22H120002）；浙江省医药卫生科技项目（2022RC069，2023KY1140）；宁波市自然科学基金（2023J390）；宁波市医药卫生科技攻关项目（2023030716） 【作者简介】蒋杰伟，博士，副教授，硕士生导师，研究方向：深度学习、眼科疾病诊断、光电图像处理、工业智能，E-mail: jiangjw924@126.com
更新日期/Last Update: 2025-09-30