A novel multi-scale network based on class attention for diabetes retinopathy

Hong Yu Chen; Rong Hua Wu; Chen Tao; Wen Jing Xu; Hui Yu; Hong Zhe Liu; Cheng Xu; Mu Wei Jian

doi:10.1109/SWC57546.2023.10448864

A novel multi-scale network based on class attention for diabetes retinopathy

Hong Yu Chen, Rong Hua Wu, Chen Tao, Wen Jing Xu, Hui Yu, Hong Zhe Liu, Cheng Xu, Mu Wei Jian^*

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

25 Downloads (Pure)

Abstract

Diabetes Retinopathy (DR) is a common eye disease, which brings irreversible blindness risk to patients in severe cases. Due to the scarcity of professional ophthalmologists, developing computer-aided diagnostic systems to participate in DR grading diagnosis has become increasingly important. However, the current mainstream deep learning methods still cannot accurately classify the severity of DR, and their unreliable results are difficult to serve as a reference for clinicians. To tackle this problem, we propose two novel modules to improve the accuracy of DR classification. Specifically, we designed a multi-scale feature extraction module (MFEM) to capture tiny lesions in fundus images and differentiate similar lesions simultaneously. In addition, we also created a class attention module (CAM) to alleviate the adverse impact of intra-class similarity on DR grading. Experiment on the APTOS2019 blind detection dataset show that our proposed two modules have made significant improvements to the designed model, achieving state-of-the-art performance with 95.98% for ACC and 97.12% for QWK, respectively.

Original language	English
Title of host publication	Proceedings - 2023 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Autonomous and Trusted Vehicles, Scalable Computing and Communications, Digital Twin, Privacy Computing and Data Security, Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PCDS/Metaverse 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
Number of pages	4
ISBN (Electronic)	9798350319804
ISBN (Print)	9798350319811
DOIs	https://doi.org/10.1109/SWC57546.2023.10448864
Publication status	Published - 1 Mar 2024
Event	9th IEEE Smart World Congress, SWC 2023 - Portsmouth, United Kingdom Duration: 28 Aug 2023 → 31 Aug 2023

Conference

Conference	9th IEEE Smart World Congress, SWC 2023
Country/Territory	United Kingdom
City	Portsmouth
Period	28/08/23 → 31/08/23

Keywords

Attention mechanism
DR grading
Fundus images
Multi-scale

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1109/SWC57546.2023.10448864

A_Novel_Multi-Scale_Network_based_on_Class_Attention_for_Diabetes_Retinopathy
© © 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Accepted author manuscript (Post-print), 1.4 MB

Cite this

Chen, H. Y., Wu, R. H., Tao, C., Xu, W. J., Yu, H., Liu, H. Z., Xu, C., & Jian, M. W. (2024). A novel multi-scale network based on class attention for diabetes retinopathy. In Proceedings - 2023 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Autonomous and Trusted Vehicles, Scalable Computing and Communications, Digital Twin, Privacy Computing and Data Security, Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PCDS/Metaverse 2023 Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SWC57546.2023.10448864

Chen, Hong Yu ; Wu, Rong Hua ; Tao, Chen et al. / A novel multi-scale network based on class attention for diabetes retinopathy. Proceedings - 2023 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Autonomous and Trusted Vehicles, Scalable Computing and Communications, Digital Twin, Privacy Computing and Data Security, Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PCDS/Metaverse 2023. Institute of Electrical and Electronics Engineers Inc., 2024.

@inproceedings{fc7dbdf21e3f44f3a98661805830b6de,

title = "A novel multi-scale network based on class attention for diabetes retinopathy",

abstract = "Diabetes Retinopathy (DR) is a common eye disease, which brings irreversible blindness risk to patients in severe cases. Due to the scarcity of professional ophthalmologists, developing computer-aided diagnostic systems to participate in DR grading diagnosis has become increasingly important. However, the current mainstream deep learning methods still cannot accurately classify the severity of DR, and their unreliable results are difficult to serve as a reference for clinicians. To tackle this problem, we propose two novel modules to improve the accuracy of DR classification. Specifically, we designed a multi-scale feature extraction module (MFEM) to capture tiny lesions in fundus images and differentiate similar lesions simultaneously. In addition, we also created a class attention module (CAM) to alleviate the adverse impact of intra-class similarity on DR grading. Experiment on the APTOS2019 blind detection dataset show that our proposed two modules have made significant improvements to the designed model, achieving state-of-the-art performance with 95.98% for ACC and 97.12% for QWK, respectively.",

keywords = "Attention mechanism, DR grading, Fundus images, Multi-scale",

author = "Chen, {Hong Yu} and Wu, {Rong Hua} and Chen Tao and Xu, {Wen Jing} and Hui Yu and Liu, {Hong Zhe} and Cheng Xu and Jian, {Mu Wei}",

note = "No ISSN; 9th IEEE Smart World Congress, SWC 2023 ; Conference date: 28-08-2023 Through 31-08-2023",

year = "2024",

month = mar,

day = "1",

doi = "10.1109/SWC57546.2023.10448864",

language = "English",

isbn = "9798350319811",

booktitle = "Proceedings - 2023 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Autonomous and Trusted Vehicles, Scalable Computing and Communications, Digital Twin, Privacy Computing and Data Security, Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PCDS/Metaverse 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

}

Chen, HY, Wu, RH, Tao, C, Xu, WJ, Yu, H, Liu, HZ, Xu, C & Jian, MW 2024, A novel multi-scale network based on class attention for diabetes retinopathy. in Proceedings - 2023 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Autonomous and Trusted Vehicles, Scalable Computing and Communications, Digital Twin, Privacy Computing and Data Security, Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PCDS/Metaverse 2023. Institute of Electrical and Electronics Engineers Inc., 9th IEEE Smart World Congress, SWC 2023, Portsmouth, United Kingdom, 28/08/23. https://doi.org/10.1109/SWC57546.2023.10448864

A novel multi-scale network based on class attention for diabetes retinopathy. / Chen, Hong Yu; Wu, Rong Hua; Tao, Chen et al.
Proceedings - 2023 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Autonomous and Trusted Vehicles, Scalable Computing and Communications, Digital Twin, Privacy Computing and Data Security, Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PCDS/Metaverse 2023. Institute of Electrical and Electronics Engineers Inc., 2024.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - A novel multi-scale network based on class attention for diabetes retinopathy

AU - Chen, Hong Yu

AU - Wu, Rong Hua

AU - Tao, Chen

AU - Xu, Wen Jing

AU - Yu, Hui

AU - Liu, Hong Zhe

AU - Xu, Cheng

AU - Jian, Mu Wei

N1 - No ISSN

PY - 2024/3/1

Y1 - 2024/3/1

N2 - Diabetes Retinopathy (DR) is a common eye disease, which brings irreversible blindness risk to patients in severe cases. Due to the scarcity of professional ophthalmologists, developing computer-aided diagnostic systems to participate in DR grading diagnosis has become increasingly important. However, the current mainstream deep learning methods still cannot accurately classify the severity of DR, and their unreliable results are difficult to serve as a reference for clinicians. To tackle this problem, we propose two novel modules to improve the accuracy of DR classification. Specifically, we designed a multi-scale feature extraction module (MFEM) to capture tiny lesions in fundus images and differentiate similar lesions simultaneously. In addition, we also created a class attention module (CAM) to alleviate the adverse impact of intra-class similarity on DR grading. Experiment on the APTOS2019 blind detection dataset show that our proposed two modules have made significant improvements to the designed model, achieving state-of-the-art performance with 95.98% for ACC and 97.12% for QWK, respectively.

AB - Diabetes Retinopathy (DR) is a common eye disease, which brings irreversible blindness risk to patients in severe cases. Due to the scarcity of professional ophthalmologists, developing computer-aided diagnostic systems to participate in DR grading diagnosis has become increasingly important. However, the current mainstream deep learning methods still cannot accurately classify the severity of DR, and their unreliable results are difficult to serve as a reference for clinicians. To tackle this problem, we propose two novel modules to improve the accuracy of DR classification. Specifically, we designed a multi-scale feature extraction module (MFEM) to capture tiny lesions in fundus images and differentiate similar lesions simultaneously. In addition, we also created a class attention module (CAM) to alleviate the adverse impact of intra-class similarity on DR grading. Experiment on the APTOS2019 blind detection dataset show that our proposed two modules have made significant improvements to the designed model, achieving state-of-the-art performance with 95.98% for ACC and 97.12% for QWK, respectively.

KW - Attention mechanism

KW - DR grading

KW - Fundus images

KW - Multi-scale

UR - http://www.scopus.com/inward/record.url?scp=85187353333&partnerID=8YFLogxK

U2 - 10.1109/SWC57546.2023.10448864

DO - 10.1109/SWC57546.2023.10448864

M3 - Conference contribution

AN - SCOPUS:85187353333

SN - 9798350319811

BT - Proceedings - 2023 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Autonomous and Trusted Vehicles, Scalable Computing and Communications, Digital Twin, Privacy Computing and Data Security, Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PCDS/Metaverse 2023

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 9th IEEE Smart World Congress, SWC 2023

Y2 - 28 August 2023 through 31 August 2023

ER -

Chen HY, Wu RH, Tao C, Xu WJ, Yu H, Liu HZ et al. A novel multi-scale network based on class attention for diabetes retinopathy. In Proceedings - 2023 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Autonomous and Trusted Vehicles, Scalable Computing and Communications, Digital Twin, Privacy Computing and Data Security, Metaverse, SmartWorld/UIC/ATC/ScalCom/DigitalTwin/PCDS/Metaverse 2023. Institute of Electrical and Electronics Engineers Inc. 2024 doi: 10.1109/SWC57546.2023.10448864

A novel multi-scale network based on class attention for diabetes retinopathy

Abstract

Conference

Keywords

UN SDGs

Access to Document

Fingerprint

Cite this