A Deep Bottleneck U-Net Combined With Saliency Map For Classifying Diabetic Retinopathy In Fundus Images

Authors

  • Vo Thi Hong Tuyet Faculty of Information Technology, Ho Chi Minh City Open University, Vietnam
  • Nguyen Thanh Binh Faculty of Computer Science and Engineering, Ho Chi Minh City University of Technology (HCMUT), Vietnam National University Ho Chi Minh City, Vietnam
  • Dang Thanh Tin Information Systems Engineering Laboratory, Faculty of Electrical and Electronics Engineering, Ho Chi Minh City University of Technology (HCMUT)- Vietnam National University Ho Chi Minh City, Vietnam

DOI:

https://doi.org/10.3991/ijoe.v18i02.27605

Keywords:

bottleneck U-net, saliency, classification, diabetic retinopathy, retinal blood vessel

Abstract


Early detection of retinopathy plays an important role in the care of people with diabetes. Classification of diabetic retinopathy in fundus images is very challenging because the blood vessels in the retinal images are too small. Morphology of objects with multi-level saliency is the recent choice because of the activation of feature extraction. However, the challenges of the input models are very complex with the blood. The color, lighting or context can become the reasons that create the decline of the primary key for training. This paper proposes a method for classification of diabetic retinopathy using saliency and shape detection of objects based on a deep Bottleneck U-Net (DbU-Net) and support vector machines  in retinal blood vessels. The proposed method includes four stages: preprocessing, feature extraction using DbU-Net, saliency prediction and classification based on the support vector machine. To evaluate this method, its results are compared to the results of the other methods by using the same datasets of STARE and DRIVE for testing with evaluation criteria such as sensitivity, specificity, and accuracy. The accuracy of the proposed method is about 97.1% in these datasets. To assess the levels of diabetes, the diagnostician must initially identify the retinal image with diabetes or not. The result of this paper may help the diagnostician to easily do this.

Author Biographies

Vo Thi Hong Tuyet, Faculty of Information Technology, Ho Chi Minh City Open University, Vietnam

Vo Thi Hong Tuyet received the Bachelor of Science degree in computer science from Ho Chi Minh City University of Pedagogical - Vietnam and the Master of Technology degree in computer science from Ho Chi Minh city University of Technology – Vietnam National University in Ho Chi Minh city (VNU-HCM), in 2011 and 2015, respectively. Now, she is a lecturer at the Faculty of Information Technology, Ho Chi Minh City Open University, Vietnam. She is working as a PhD student at Faculty of Computer science and Engineering, the Ho Chi Minh City University of Technology, VNU-HCM. Her research interests include recognition, image processing.

Nguyen Thanh Binh, Faculty of Computer Science and Engineering, Ho Chi Minh City University of Technology (HCMUT), Vietnam National University Ho Chi Minh City, Vietnam

Nguyen Thanh Binh received the Bachelor of Engineering degree from Ho Chi Minh City University of Technology -Vietnam National University Ho Chi Minh City (VNU-HCM), Viet Nam, in 2000, the Master's degree and Ph.D degree in computer science both from University of Allahabad, India, in 2005 and 2011, respectively. Now, he is currently an Associate Professor in the Faculty of Computer science and Engineering, the Ho Chi Minh City University of Technology, VNU-HCM. He has published one book, one book chapter and more than 65 research papers. His research interests include recognition, image processing, multimedia information systems, decision support system, and time series data.

Dang Thanh Tin, Information Systems Engineering Laboratory, Faculty of Electrical and Electronics Engineering, Ho Chi Minh City University of Technology (HCMUT)- Vietnam National University Ho Chi Minh City, Vietnam

Dang Thanh Tin received his M.S. degree in Electronic Engineering from the Ho Chi Minh city University of Technology, Viet Nam, and his Ph.D degree in Information system engineering in the cooperation of the Ho Chi Minh city University of Technology and Osaka Sangyo University, Japan, in 1998 and in 2006, respectively. From 2007 to 2013, he was at the Ho Chi Minh city University of Technology as Assistant Professor. He is currently an Associate Professor in the Faculty of Electrical and Electronics Engineering, the Ho Chi Minh city University of Technology, Viet Nam. In 2006, 2009, and 2016 he was a research fellow, and visiting professor in Faculty of Electrical and Computer Engineering at University of Illinois at Urbana Champaign (USA), and in Lab of LCS (Laboratoire de Conception et d'Intégration des Systèmes) at Grenoble INP-Esisar (FRANCE), where he was responsible for research programs in the area of Computer systems engineering and Information systems engineering. Dang Thanh Tin is the author of over 52 technical publications. His research interests include image processing topics in biomedical engineering, medical expert systems topics.

Downloads

Published

2022-02-16

How to Cite

Tuyet, V. T. H., Binh, N. T., & Tin, D. T. (2022). A Deep Bottleneck U-Net Combined With Saliency Map For Classifying Diabetic Retinopathy In Fundus Images. International Journal of Online and Biomedical Engineering (iJOE), 18(02), pp. 105–122. https://doi.org/10.3991/ijoe.v18i02.27605

Issue

Section

Papers