Diabetic Retinopathy Detection Using Genetic Algorithm-Based CNN Features and Error Correction Output Code SVM Framework Classification Model

Wireless Communications and Mobile Computing - Tập 2022 - Trang 1-13 - 2022
Najib Ullah1, Muhammad Ismail Mohmand1, Kifayat Ullah2, Mohammed S. M. Gismalla3, Liaqat Ali4, Shafqat Ullah Khan5, Niamat Ullah6
1Department of Computer Science, The Brains Institute Peshawar, Pakistan
2Department of Computer and Software Technology, University of Swat, Pakistan
3Department of Electronic and Electrical Engineering, Faculty of Engineering, International University of Africa, Khartoum, Sudan
4Department of Electrical Engineering, University of Science and Technology, Bannu, Pakistan
5Department of Electronics, University of Buner, Pakistan
6Department of Computer Science, University of Buner, Pakistan

Tóm tắt

Diabetic retinopathy (DR) is a type of eye disease that may be caused in individuals suffering from diabetes which results in vision loss. DR identification and routine diagnosis is a challenging task and may need several screenings. Early identification of DR has the potential to prevent or delay vision loss. For real-time applications, an automated DR identification approach is required to assist and reduce possible human mistakes. In this research work, we propose a deep neural network and genetic algorithm-based feature selection approach. Five advanced convolutional neural network architectures are used to extract features from the fundus images, i.e., AlexNet, NASNet-Large, VGG-19, Inception V3, and ShuffleNet, followed by genetic algorithm for feature selection and ranking features into high rank (optimal) and lower rank (unsatisfactory). The nonoptimal feature attributes from the training and validation feature vectors are then dropped. Support vector machine- (SVM-) based classification model is used to develop diabetic retinopathy recognition model. The model performance is evaluated using accuracy, precision, recall, and F1 score. The proposed model is tested on three different datasets: the Kaggle dataset, a self-generated custom dataset, and an enhanced custom dataset with 97.9%, 94.76%, and 96.4% accuracy, respectively. In the enhanced custom dataset, data augmentation has been performed due to the smaller size of the dataset and to eliminate the noise in fundus images.

Từ khóa


Tài liệu tham khảo

10.1016/j.ophtha.2017.02.008

M. Chetoui, Diabetic retinopathy detection using machine learning and texture features, 1

10.1109/ACCESS.2020.2980055

D. K. Prasad, Early detection of diabetic retinopathy from digital retinal fundus images, 240

10.1109/ACCESS.2019.2903171

10.1007/s00521-018-03974-0

10.1109/ACCESS.2018.2888639

P. Junjun, Diabetic retinopathy detection based on deep convolutional neural networks for localization of discriminative regions, 46

10.1016/j.patrec.2020.04.026

R. Ghosh, Automatic detection and classification of diabetic retinopathy stages using CNN, 10.1109/SPIN.2017.8050011

10.1371/journal.pone.0179790

10.1016/j.procs.2016.07.014

10.1001/jama.2016.17216

10.1016/j.ins.2007.07.020

Y. S. Kanungo, Detecting diabetic retinopathy using deep learning, 801

S. Masood, Identification of Diabetic Retinopathy in Eye Images Using Transfer Learning, 10.1109/CCAA.2017.8229977

A. Kwasigroch, Deep CNN based decision support system for detection and assessing the stage of diabetic retinopathy, 10.1109/IIPHDW.2018.8388337

X. Wang, Diabetic retinopathy stage classification using convolutional neural networks, 10.1109/IRI.2018.00074

G. Garc, Detection of diabetic retinopathy based on a convolutional neural network using retinal fundus images, 1

10.1038/s41746-019-0099-8

10.1109/ACCESS.2019.2947484

I. Ardiyanto, Deep learning-based diabetic retinopathy assessment on embedded system, 1760

Grissworld Home Care, The 4 stages of diabetic retinopathy: what you can expect

10.3390/sym11010001

10.1016/j.comcom.2017.10.005

10.3390/e20040290

10.1109/ACCESS.2021.3074422

10.1016/j.neucom.2019.06.084

Thông tin
Thông tin xuất bản

Wireless Communications and Mobile Computing

Tập 2022

1-13

Thông tin tác giả