A Deep Feature Fusion of Improved Suspected Keratoconus Detection with Deep Learning

被引：16

作者：

Al-Timemy, Ali H. ^{[1
]}

Alzubaidi, Laith ^{[2
,3
]}

Mosa, Zahraa M. ^{[4
]}

Abdelmotaal, Hazem ^{[5
]}

Ghaeb, Nebras H. ^{[1
]}

Lavric, Alexandru ^{[6
]}

Hazarbassanov, Rossen M. ^{[7
,8
]}

Takahashi, Hidenori ^{[9
]}

Gu, Yuantong ^{[2
,3
]}

Yousefi, Siamak ^{[10
,11
]}

机构：

[1] Univ Baghdad, Al Khwarizmi Coll Engn, Biomed Engn Dept, Baghdad, Iraq

[2] Queensland Univ Technol, Sch Mech Med & Proc Engn, Brisbane, Qld 4000, Australia

[3] Queensland Univ Technol, ARC Ind Transformat Training Ctr, Joint Biomech, Brisbane, Qld 4000, Australia

[4] Al Nahrain Univ, Coll Sci, Dept Phys, Baghdad, Iraq

[5] Assiut Univ, Dept Ophthalmol, Assiut 71526, Egypt

[6] Stefan Cel Mare Univ Suceava, Comp Elect & Automat Dept, Suceava 720229, Romania

[7] Univ Anhembi Morumbi, Med Sch, BR-03101001 Sao Paulo, Brazil

[8] Univ Fed Sao Paulo, Paulista Med Sch, Dept Ophthalmol & Visual Sci, BR-04021001 Sao Paulo, Brazil

[9] Jichi Med Univ, Dept Ophthalmol, Shimotsuke, Tochigi 3290431, Japan

[10] Univ Tennessee, Hlth Sci Ctr, Dept Ophthalmol, Memphis, TN 38163 USA

[11] Univ Tennessee, Hlth Sci Ctr, Dept Genet Genom & Informat, Memphis, TN 38163 USA

来源：

DIAGNOSTICS | 2023年 / 13卷 / 10期

关键词：

convolutional neural networks; keratoconus; feature fusion; transfer learning; deep learning; machine learning; CORNEAL TOPOGRAPHY; RAW DATA; CLASSIFICATION;

D O I：

10.3390/diagnostics13101689

中图分类号：

R5 [内科学];

学科分类号：

1002 ; 100201 ;

摘要：

Detection of early clinical keratoconus (KCN) is a challenging task, even for expert clinicians. In this study, we propose a deep learning (DL) model to address this challenge. We first used Xception and InceptionResNetV2 DL architectures to extract features from three different corneal maps collected from 1371 eyes examined in an eye clinic in Egypt. We then fused features using Xception and InceptionResNetV2 to detect subclinical forms of KCN more accurately and robustly. We obtained an area under the receiver operating characteristic curves (AUC) of 0.99 and an accuracy range of 97-100% to distinguish normal eyes from eyes with subclinical and established KCN. We further validated the model based on an independent dataset with 213 eyes examined in Iraq and obtained AUCs of 0.91-0.92 and an accuracy range of 88-92%. The proposed model is a step toward improving the detection of clinical and subclinical forms of KCN.

引用

页数：13

共 30 条

[1] Classification of Color-Coded Scheimpflug Camera Corneal Tomography Images Using Deep Learning [J].

Abdelmotaal, Hazem ;

Mostafa, Magdi M. ;

Mostafa, Ali N. R. ;

Mohamed, Abdelsalam A. ;

Abdelazeem, Khaled .

TRANSLATIONAL VISION SCIENCE & TECHNOLOGY, 2020, 9 (13) :1-12

[2]

Al-Timemy A, 2022, Investigative Ophthalmology and Visual Science, V63, P2101

[3] A Hybrid Deep Learning Construct for Detecting Keratoconus From Corneal Maps [J].

Al-Timemy, Ali H. ;

Mosa, Zahraa M. ;

Alyasseri, Zaid ;

Lavric, Alexandru ;

Lui, Marcelo M. ;

Hazarbassanov, Rossen M. ;

Yousefi, Siamak .

TRANSLATIONAL VISION SCIENCE & TECHNOLOGY, 2021, 10 (14)

[4] Deep Transfer Learning for Improved Detection of Keratoconus using Corneal Topographic Maps [J].

Al-Timemy, Ali H. ;

Ghaeb, Nebras H. ;

Mosa, Zahraa M. ;

Escudero, Javier .

COGNITIVE COMPUTATION, 2022, 14 (05) :1627-1642

[5] A systematic review of trustworthy and explainable artificial intelligence in healthcare: Assessment of quality, bias risk, and data fusion [J].

Albahri, A. S. ;

Duhaim, Ali M. ;

Fadhel, Mohammed A. ;

Alnoor, Alhamzah ;

Baqer, Noor S. ;

Alzubaidi, Laith ;

Albahri, O. S. ;

Alamoodi, A. H. ;

Bai, Jinshuai ;

Salhi, Asma ;

Santamaria, Jose ;

Ouyang, Chun ;

Gupta, Ashish ;

Gu, Yuantong ;

Deveci, Muhammet .

INFORMATION FUSION, 2023, 96 :156-191

[6] A Hybrid Artificial Intelligence Model for Detecting Keratoconus [J].

Alyasseri, Zaid Abdi Alkareem ;

Al-Timemy, Ali H. ;

Abasi, Ammar Kamal ;

Lavric, Alexandru ;

Mohammed, Husam Jasim ;

Takahashi, Hidenori ;

Milhomens Filho, Jose Arthur ;

Campos, Mauro ;

Hazarbassanov, Rossen M. ;

Yousefi, Siamak .

APPLIED SCIENCES-BASEL, 2022, 12 (24)

[7] A survey on deep learning tools dealing with data scarcity: definitions, challenges, solutions, tips, and applications [J].

Alzubaidi, Laith ;

Bai, Jinshuai ;

Al-Sabaawi, Aiman ;

Santamaria, Jose ;

Albahri, A. S. ;

Al-dabbagh, Bashar Sami Nayyef ;

Fadhel, Mohammed. A. A. ;

Manoufali, Mohamed ;

Zhang, Jinglan ;

Al-Timemy, Ali. H. H. ;

Duan, Ye ;

Abdullah, Amjed ;

Farhan, Laith ;

Lu, Yi ;

Gupta, Ashish ;

Albu, Felix ;

Abbosh, Amin ;

Gu, Yuantong .

JOURNAL OF BIG DATA, 2023, 10 (01)

[8] Review of deep learning: concepts, CNN architectures, challenges, applications, future directions [J].

Alzubaidi, Laith ;

Zhang, Jinglan ;

Humaidi, Amjad J. ;

Al-Dujaili, Ayad ;

Duan, Ye ;

Al-Shamma, Omran ;

Santamaria, J. ;

Fadhel, Mohammed A. ;

Al-Amidie, Muthana ;

Farhan, Laith .

JOURNAL OF BIG DATA, 2021, 8 (01)

[9] Keratoconus detection of changes using deep learning of colour-coded maps [J].

Chen, Xu ;

Zhao, Jiaxin ;

Iselin, Katja C. ;

Borroni, Davide ;

Romano, Davide ;

Gokul, Akilesh ;

McGhee, Charles N. J. ;

Zhao, Yitian ;

Sedaghat, Mohammad-Reza ;

Momeni-Moghaddam, Hamed ;

Ziaei, Mohammed ;

Kaye, Stephen ;

Romano, Vito ;

Zheng, Yalin .

BMJ OPEN OPHTHALMOLOGY, 2021, 6 (01)

[10] Xception: Deep Learning with Depthwise Separable Convolutions [J].

Chollet, Francois .

30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :1800-1807

← 1 2 3 →