Diabetic retinopathy screening using deep learning for multi-class imbalanced datasets. (October 2022)

• Record Type: Journal Article
• Authors: Saini, Manisha; Susan, Seba
• Is Part Of: Computers in biology and medicine. Volume 149(2022)
• Publication Date: 2022-10
• Subjects: Diabetic retinopathy -- Deep learning -- Image classification -- Object detection -- Segmentation -- Transfer learning; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2022.105989
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: Screening and diagnosis of diabetic retinopathy disease is a well known problem in the biomedical domain. The use of medical imagery from a patient's eye for detecting the damage caused to blood vessels is a part of the computer-aided diagnosis that has immensely progressed over the past few years due to the advent and success of deep learning. The challenges related to imbalanced datasets, inconsistent annotations, less number of sample images and inappropriate performance evaluation metrics has caused an adverse impact on the performance of the deep learning models. In order to tackle the effect caused by class imbalance, we have done extensive comparative analysis between various state-of-the-art methods on three benchmark datasets of diabetic retinopathy: - Kaggle DR detection, IDRiD and DDR, for classification, object detection and segmentation tasks. This research could serve as a concrete baseline for future research in this field to find appropriate approaches and deep learning architectures for imbalanced datasets. Highlights: Extensive comparative analysis between various state-of-the-art methods involving transfer learning on three benchmark datasets of varied sizes consisting of diabetic retinopathy images for DR grading (classification), segmentation, detection of lesions and optical disc. Application of different pre-trained CNN architectures to fundus images combined with rejection resampling (random under-sampling at mini-batch level) technique for effective performance in tackling different imbalanced scenarios of varied dataset sizes. Automated system for diabetic retinopathy screening using classification, object detection and segmentation of fundus images as a unified approach for different computer vision tasks for application in the biomedical domain. Pre-processing of images using automated RandAugment approach for effective data augmentation for enhancing classification tasks in the biomedical domain apart from other augmentation techniques used for regularization of segmentation and object detection tasks.