实用老年医学

实用老年医学 ›› 2022, Vol. 36 ›› Issue (3): 280-283.doi: 10.3969/j.issn.1003-9198.2022.03.017

• 临床研究 • 上一篇    下一篇

基于人工智能ResNeXt的高度近视诊断方法

万程, 陈柏兵, 沈建新, 陈志强   

  1. 210016 江苏省南京市,南京航空航天大学电子信息工程学院(万程,陈柏兵,沈建新);
    210024 江苏省南京市,江苏省省级机关医院眼科(陈志强)
  • 收稿日期:2021-06-20 出版日期:2022-03-20 发布日期:2022-03-29
  • 通讯作者: 陈志强,Email:chenzhiqiangmail@163.com
  • 基金资助:
    中国博士后科学基金资助项目(2019M661832);江苏省博士后科研资助计划(2019K226);江苏高校优势学科建设工程项目

A diagnostic method of high myopia based on artificial intelligence ResNeXt

WAN Cheng, CHEN Bai-bing, SHEN Jian-xin, CHEN Zhi-qiang   

  1. WAN Cheng, CHEN Bai-bing, SHEN Jian-xin. College of Electronic Information Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China;
    CHEN Zhi-qiang. Department of Ophthalmology, Jiangsu Province Geriatric Hospital, Nanjing 210024, China
  • Received:2021-06-20 Online:2022-03-20 Published:2022-03-29

PDF (PC)

1317

摘要： 目的 采用人工智能诊断方法,提高高度近视诊断效率,辅助医疗工作者诊断。 方法 使用参数量少、训练速度快的深度学习网络ResNeXt-50进行高度近视诊断,区分正常眼底与高度近视眼底。 结果 本文采用了江苏省省级机关医院的6571张高度近视彩色照片和6212张正常彩色照片作为数据集。ResNeXt-50在包含2558张眼底图像的测试集上取得了94.10%的准确度、92.33%的灵敏度、95.94%的特异度,AUC为0.9861。平均每张图片耗时0.035 s,在实时诊断的可接受范围内,满足了医疗辅助诊断的实时性。 结论 ResNeXt-50网络能够高效、准确地诊断高度近视。

关键词: 高度近视, 计算机辅助诊断, 卷积神经网络, 图像分类

Abstract: Objective To promote the development of computer-aided diagnosis, and to improve the diagnosing efficiency of high myopia. Methods ResNeXt-50 network was used to diagnose high myopia with few parameters and fast training speed. It was used to distinguish normal fundus and high myopia fundus in this study. Results This study used 6571 high myopia color photos and 6212 normal color photos from Jiangsu Province Geriatric Hospital as the data set. In the end, the diagnostic method got an accuracy of 94.10%, a sensitivity of 92.33%, a specificity of 95.94% and an AUC of 0.9861 for high myopia. The average time of each image took 0.035 s, which was within the acceptable range of real-time diagnosis, which met the real-time performance of medical auxiliary diagnosis. Conclusions ResNeXt-50 has a good classification performance. It can diagnose high myopia efficiently and accurately.

Key words: high myopia, computer-aided diagnosis, convolutional neural network, image classification

中图分类号: