Automated identification of cataract severity using retinal fundus images

ABSTRACT

Cataract is the most prevalent cause of blindness worldwide, which accounts for more than 51% of overall blindness. The early detection of cataract can salvage impaired vision leading to blindness. Most of the existing cataract classification systems are based on traditional machine learning methods with hand-engineered features. The manual extraction of retinal features is generally a time-taking process and requires professional ophthalmologists. Convolutional neural network (CNN) is a widely accepted model for automatic feature extraction, but it necessitates a larger dataset to evade overfitting problems. Contrarily, classification using SVM has great generalisation power to elucidate small-sample dataset. Therefore, we proposed a hybrid model by integrating deep learning models and SVM for 4-class cataract classification. The transfer learning-based models (AlexNet, VGGNet, ResNet) are employed for automatic feature extraction and SVM performs as a recogniser. The proposed architecture evaluated on 8030 retinal images with strong feature extraction and classification techniques has achieved 95.65% of accuracy. The results of this study have verified that the proposed method outperforms conventional methods and can provide a reference for other retinal diseases.

KEYWORDS:

• Convolutional neural network
• SVM
• cataract detection
• retinal images
• image classification

Disclosure statement

The authors declare no conflict of interest.Figure 6

Additional information

Funding

This work is supported by the National Natural Science Foundation of China with project no. [81970844].

Notes on contributors

Azhar Imran

AZHAR IMRAN is currently pursuing Ph.D. degree from Beijing University of Technology, Beijing, China. His research interests include machine learning, image processing, medical imaging, and data mining.

Jianqiang Li

JIANQIANG LI received Ph.D. degrees in control science and engineering from Tsinghua University, Beijing, in 2004. He joined the Beijing University of Technology and the Beijing Engineering Research Center for IoT Software and Systems, Beijing, in 2013, as a Beijing Distinguished Professor.  His research interests include Petri nets, enterprise information systems, and big data.

Yan Pei

YAN PEI received the Dr.Eng. degree from Kyushu University, Fukuoka, Japan, He is currently an Associate Professor with the University of Aizu. His research interests include evolutionary computation, machine learning, and software engineering.

Faheem Akhtar

FAHEEM AKHTAR is currently an Assistant Professor with the Department of Computer Science, Sukkur IBA Univeristy. Meanwhile, he is on study leave from Sukkur IBA to pursue his Ph.D. degree with the School of Software Engineering, Beijing University of Technology, Beijing, China. His research interests include data mining, machine learning, and big data.

Ji-Jiang Yang

JI-JIANG YANG received the Ph.D. degree from the National University of Ireland, Galway. He is currently an Associate Professor with Tsinghua University. His research interests include e-health, e-government/e-commerce, privacy-preserving, information resource management, data mining, and cloud computing.

Yanping Dang

YANPING DANG received her BS degree in Clinical Medicine Scince from Tianjin Medical University, China in 1997. She worked as Physician in the Beijinhg Moslem People's Hospitl until now. Her Research Interests include General Medicine.