An Improved Method for Automated Identification of Hard Exudates in Diabetic Retinopathy Disease

Abstract

The presence of Hard Exudates in retinal images is one of the significant indications of Diabetic Retinopathy disease. Early detection of Hard Exudates obviously prevents the vision loss very effectively. Today, the development of appropriate Pre-processing stages for Hard Exudate identification through automated screening is an open issue. In this research article, a mathematical morphology based image processing technique is presented for the identification of Hard Exudates in retinal images. The analysis of noisy and low-contrast retinal images shows the importance of this method. In this research work, 570 input images collected from Hospital are examined by the k-fold cross-validation technique. The sensitivity and specificity, two important parameters for performance analysis of Hard Exudate screening, are 97.36% and 99.31%, respectively. In Receiver Operating Characteristics the Area Under Curve (AUC) is 0.93 (average for k folds). Public datasets e_Ophtha_EX and DIARETDB1 are used here to compare the screening performance with existing algorithms and AUC is 0.97 and 0.98, respectively. Thus, the screening result on both Hospital dataset and Public dataset reflects the effectiveness of the proposed method.

KEYWORDS:

• Contrast enhancement
• Diabetic retinopathy screening system
• Hard Exudates
• k-fold cross-validation

Acknowledgements

The authors would like to thank The Ophthalmology Department of Sri Aurobindo Seva Kendra, Kolkata, India, for their clinical support.

Additional information

Funding

This research activity is financially supported by WB DST-FIST, Memo No: 1203(Sanc.)/ST/P/S&T/Misc-9/14 and R & D Project, Memo No: 148(Sanc.)/ST/P/S&T/6G-13/2018 of Science & Technology and Biotechnology Dept., Government of West Bengal, India.

Notes on contributors

Niladri Sekhar Datta

Niladri Sekhar Datta obtained BTech in CSE from West Bengal University of Technology, India. He obtained MTech from the University of Calcutta, India. His research interests include medical imaging and pattern recognition. Corresponding author. E-mail: [email protected]

Himadri Sekhar Dutta

Himadri Sekhar Dutta obtained BTech in ECE from University of Kalyani, India. He obtained MTech and PhD from the University of Calcutta. He published more than 35 research articles in reputed journals and conferences. His research interests include medical image processing, embedded systems and opto electronics devices. E-mail: [email protected]

Koushik Majumder

Koushik Majumder obtained BTech and MTech from the University of Calcutta and PhD from Jadavpur University. His research interests include medical image processing and mobile computing. He published several research articles in reputed international conferences and journals. E-mail: [email protected]

Sumana Chatterjee

Sumana Chatterjee obtained MBBS from Bardawan Medical College. She obtained MS from West Bengal Medical University. She completed FMRF from Shankar Netralaya, India. Her research interests include cornea and ocular surface of human eye. E-mail: [email protected]

Najir Abdul Wasim

Nazir Abdul Wasim obtained MBBS from Burwan Medical College and Hospital. His research interest includes medical data analysis. E-mail: [email protected]