当前位置:网站首页>Selected review | machine learning technology for Cataract Classification / classification
Selected review | machine learning technology for Cataract Classification / classification
2022-07-04 20:26:00 【Zhiyuan community】
【MIR Reading guide 】 Worldwide , Cataract is the main cause of visual impairment and blindness . these years , Researchers have made great progress in developing the most advanced machine learning technology for automatic Cataract Classification and classification , Aimed at early prevention of cataract , Improve the diagnostic efficiency of clinicians . The research team from Southern University of science and technology comprehensively reviewed Cataract Classification Based on ophthalmic images / The latest development of hierarchical machine learning technology . This paper summarizes the existing literature from the two directions of traditional machine learning methods and deep learning methods , And in-depth analysis of the advantages and limitations of the existing research . Besides , The article also discusses the automatic Cataract Classification Based on machine learning technology / Some challenges faced by grading technology , And put forward possible solutions for future research .
According to the World Health Organization , Around the world 22 Billion people suffer from visual impairment . Cataracts account for about% of visual impairment 33%, It is the number one cause of blindness worldwide ( exceed 50%). Cataract patients can improve their quality of life and vision through early intervention and cataract surgery , This is to reduce the blindness rate at the same time 、 An effective method to reduce the burden of blindness caused by cataract in society .
Clinically , When proteins in the crystalline body gather , The transparency of lens area decreases , And then cause cataracts . This is related to many factors , For example, dysplasia 、 Trauma 、 Metabolic disorders 、 Genetic factors, 、 Drug induced changes 、 Age etc. . Heredity and age are the two most important factors that cause cataracts .
In the last few years , Ophthalmologists based on their experience and clinical training , Use several ophthalmic images to diagnose cataracts . This manual diagnosis mode is easy to make mistakes 、 Time consuming 、 Subjective and costly , And experienced clinicians are scarce , This gives developing countries 、 rural 、 Cataract screening and diagnosis and treatment in the community bring great challenges . In order to prevent cataract in the early stage , Improve the accuracy and efficiency of cataract diagnosis , Researchers are committed to developing computer-aided diagnosis (CAD) technology , Including traditional machine learning methods and deep learning methods , For different ophthalmic images , Realize the automatic classification of cataract / classification .
In the past decade , Deep learning has achieved great success in all fields , Including medical image analysis . It can learn low-level from raw data in an end-to-end way 、 Intermediate and advanced feature representation ( for example , Ophthalmic image ). Various deep neural network models have been used to deal with Cataract Classification / Graded tasks , For example, convolutional neural networks (CNN)、 Attention based networks 、 Fast RCNN And multi-layer perceptron (MLP) neural network .
The existing review articles summarize the types of cataract 、 Cataract Classification / Grading system and ophthalmic imaging mode ; However , So far, no article has systematically summarized the automatic Cataract Classification Based on ophthalmic imaging mode / classification ML technology . This paper summarizes systematically for the first time ML Technology for automatic Cataract Classification / The latest progress in grading , Focus on Cataract Classification / In grading ML technology , Including tradition ML Methods and deep learning methods .
This paper summarizes Web of Science、Scopus and Google Scholar Related papers in the database . Based on the collected papers 、 Summary of the research team and communication with experienced ophthalmologists , It forms the overall organizational framework of this paper ( As shown in Fig. 1 ). The research team also briefly reviewed ophthalmic imaging patterns 、 Cataract grading system and common evaluation methods , And gradually introduced ML technology , In order to provide a valuable summary for the current research , And based on ML Cataract Classification / The classification points out the potential research direction in the future .
Machine Learning for Cataract Classification/Grading on Ophthalmic Imaging Modalities: A Survey
Xiao-Qing Zhang, Yan Hu, Zun-Jie Xiao, Jian-Sheng Fang, Risa Higashita, Jiang Liu
https://link.springer.com/article/10.1007/s11633-022-1329-0
https://www.mi-research.net/en/article/doi/10.1007/s11633-022-1329-0
【 The author of this article 】
About Machine Intelligence Research
Machine Intelligence Research( abbreviation MIR, Original title International Journal of Automation and Computing) Sponsored by the Institute of automation, Chinese Academy of Sciences , On 2022 It was officially published in .MIR Based on the domestic 、 Global oriented , Focus on serving the national strategic needs , Publish the latest original research papers in the field of machine intelligence 、 review 、 Comments, etc , Comprehensively report the basic theories and cutting-edge innovative research achievements in the field of international machine intelligence , Promote international academic exchanges and discipline development , Serve the progress of national artificial intelligence science and technology . The journal was selected " China Science and technology journal excellence action plan ", Has been ESCI、EI、Scopus、 The core journals of science and technology in China 、CSCD Wait for the database to include .
边栏推荐
- FS4061A升压8.4V充电IC芯片和FS4061B升压12.6V充电IC芯片规格书datasheet
- Integritee通过XCM集成至Moonriver,为其生态系统带来企业级隐私解决方案
- Actual combat simulation │ JWT login authentication
- QT writing the Internet of things management platform 38- multiple database support
- What are the consequences of closing the read / write channel?
- Kotlin cycle control
- Detailed explanation of Audi EDI invoice message
- Informatics Olympiad 1336: [example 3-1] find roots and children
- Basic use of kotlin
- 应用实践 | 蜀海供应链基于 Apache Doris 的数据中台建设
猜你喜欢
华为nova 10系列支持应用安全检测功能 筑牢手机安全防火墙
AP8022开关电源小家电ACDC芯片离线式开关电源IC
Actual combat simulation │ JWT login authentication
多表操作-外连接查询
Related concepts of federal learning and motivation (1)
实战模拟│JWT 登录认证
Neural network IOT platform construction (IOT platform construction practical tutorial)
#夏日挑战赛#带你玩转HarmonyOS多端钢琴演奏
【深度学习】一文看尽Pytorch之十九种损失函数
Application practice | Shuhai supply chain construction of data center based on Apache Doris
随机推荐
What should we pay attention to when doing social media marketing? Here is the success secret of shopline sellers!
Regular replacement [JS, regular expression]
实践示例理解js强缓存协商缓存
CANN算子:利用迭代器高效实现Tensor数据切割分块处理
凌云出海记 | 沐融科技&华为云:打造非洲金融SaaS解决方案样板
Pytoch learning (4)
[Beijing Xunwei] i.mx6ull development board porting Debian file system
Swagger suddenly went crazy
华为云云商店首页 Banner 资源位申请
Qt编写物联网管理平台38-多种数据库支持
Niuke Xiaobai month race 7 e applese's super ability
Data set division
1005 spell it right (20 points) (pat a)
YOLOv5s-ShuffleNetV2
Thinking on demand development
Introduction to ACM combination counting
水晶光电:长安深蓝SL03的AR-HUD产品由公司供应
C语言-入门-基础-语法-流程控制(七)
实战模拟│JWT 登录认证
多表操作-内连接查询