Mask Distillation Network for Conjunctival Hyperemia Severity Classification
作者机构：School of Computer Science and EngineeringNanjing University of Science and TechnologyNanjing210094China Department of OphthalmologyThe First Affiliated Hospital with Nanjing Medical UniversityNanjing210029China
出 版 物：《Machine Intelligence Research》 (机器智能研究（英文版）)
年 卷 期：2023年第20卷第6期
页 面：909-922页
核心收录：
学科分类：12[管理学] 07[理学] 08[工学] 0831[工学-生物医学工程（可授工学、理学、医学学位）] 0711[理学-系统科学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 080401[工学-精密仪器及机械] 0804[工学-仪器科学与技术] 080402[工学-测试计量技术及仪器] 0835[工学-软件工程] 0836[工学-生物工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）]
基 金：This work was supported in part by National Natural Science Foundation of China(Nos.62172223 and 61671242) the Fundamental Research Funds for the Central Universities(No.30921013105)
主 题：Mask distillation(MD) conjunctiva hyperemia attention mechanism severity classification deep learning
摘 要：To achieve automatic,fast and accurate severity classification of bulbar conjunctival hyperemia severity,we proposed a novel prior knowledge-based framework called mask distillation network(MDN).The proposed MDN consists of a segmentation network and a classification network with teacher-student *** segmentation network is used to generate a bulbar conjunctival mask and the classification network divides the severity of bulbar conjunctival hyperemia into four *** the classification network,we feed the original image and the image with the bulbar conjunctival mask into the student and teacher branches respectively,and an attention consistency loss and a classification consistency loss are used to keep a similar learning mode for these two *** design of“different input but same output,named mask distillation(MD),aims to introduce the regional prior knowledge that“bulbar conjunctival hyperemia severity classification is only related to the bulbar conjunctiva region.Extensive experiments on 5117 anterior segment images have proven the effectiveness of mask distillation technology:1)The accuracy of the MDN student branch is 3.5%higher than that of a single optimal baseline network and 2%higher than that of the baseline network combination.2)In the test phase,only the student branch is needed,and no additional segmentation network is *** framework only takes 0.003 s to classify a single image,achieving the fastest speed in all the methods we compared.3)Compared with a single baseline network,the attention of both teacher and student branches in the MDN has been intuitively improved.
维普期刊数据库