Deep Learning Techniques for Diabetic Retinopathy Classification: A Survey

Atwany, Mohammad Z.; Sahyoun, Abdulwahab; Yaqub, Mohammad

doi:10.1109/access.2022.3157632

Cited by 123 publications

(62 citation statements)

References 70 publications

(108 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From this perspective, it is seen that machine learning has always played a dominant role. This can be seen in the review presented by Atwany et al [12]. According to this study, deep learning offers a significant advantage in classification, but there is still a broad scope to improve the system's computational efficiency.…”

Section: Related Workmentioning

confidence: 81%

Novel Framework for Enhanced Learning-based Classification of Lesion in Diabetic Retinopathy

Prakruthi¹,

Komarasamy²

2022

IJACSA

View full text Add to dashboard Cite

Diabetic retinopathy is an adverse medical condition resulting from a high level of blood sugar potentially affecting the retina and leading to permanent vision loss in its advanced stage of progression. A literature review is conducted to assess the effectiveness of existing approaches to find that Convolution Neural Network (CNN) has been frequently adopted for analyzing the fundus retinal image for detection and classification. However, existing scientific methods are mainly inclined towards achieving accuracy in their learning techniques without much deeper investigation of possibilities to improve the methodology of type using CNN. Therefore, the proposed scheme introduces a computational framework where a simplified feature enhancement operation is carried out, resulting in artifact-free images with better features. The enhanced image is then subjected to CNN to perform multiclass categorization of potential stages of diabetic retinopathy to see if it outperforms existing schemes.

show abstract

Section: Related Workmentioning

confidence: 81%

Novel Framework for Enhanced Learning-based Classification of Lesion in Diabetic Retinopathy

Prakruthi¹,

Komarasamy²

2022

IJACSA

View full text Add to dashboard Cite

show abstract

“…These models motivated the researchers to develop fast and accurate computer-aided diagnosis (CAD) systems for different machine learning and image processing applications. These models are mostly used in age assessment [11], diabetic retinopathy screening [12], [13], lung cancer [14], breast cancer [15], skin cancer [16], brain tumor [17], cervical cancer [18], liver cancer [19], bone disease prediction [20].…”

Section: Introductionmentioning

confidence: 99%

MSF-Model: Multi-Scale Feature Fusion-Based Domain Adaptive Model for Breast Cancer Classification of Histopathology Images

et al. 2022

View full text Add to dashboard Cite

One of the most common causes of mortality for women globally is breast cancer. Early breast cancer identification could make it possible for people to receive the appropriate treatment to save their lives and return to their routine lives. Breast cancer diagnosis by histopathology is referred to as the gold standard. In recent years, convolutional neural network-based techniques are used for breast cancer classification. However, they faced domain adaptation, small objects retention, and feature extraction issues of complex microscopic images. In this study, we introduced multi-scale feature fusion-based domain adaptive model for breast cancer classification using histopathology images. It has two blocks and six lightweight submodels where each block contains three models. Dilated layers are used in sub-models to overcome the disappearing of small objects in deep layers. Reducing the disappearing of small objects helped to extract better features for higher performance. Multiple heterogeneous feature extractors are used in this study which helped to extract various features. Extracted features are fused and reduced by retaining better features. Learning of model from natural images to complex microscopic images has limitation of domain adaptation. Same domain transfer learning is used in this study to overcome the limitations of different domain transfer learning. Model is trained on patchcamelyon17 dataset and weights of this training are further used for same domain transfer learning. Pre-trained weights are further used for the training of proposed model on BreaKHis dataset. A number of conventional data augmentation techniques are used as complex models require higher number of samples for the tuning of weights. Local window based CLAHE contrast enhancement technique is used to increase foreground-background contrast and remove noise. The proposed model achieved 98.00% precision, 98.15% recall, 98.08% f-measure, and 98.23% accuracy on test data. To best of our knowledge, it surpassed state-of-the-art models.

show abstract

“…Given the fact that medical diagnosis is usually expensive and requires costly equipment, there is a dire need for a cheap and viable solution [3]. Computer-Aided Diagnosis using Convolutional Neural Networks (CNN) plays a key role in DR detection as the cost can be significantly reduced while making the diagnosis more accessible [4]. CNN is a Deep Learning technique where image features are extracted and classified by the model.…”

Section: Introductionmentioning

confidence: 99%

“…CNN is a Deep Learning technique where image features are extracted and classified by the model. Considerable amounts of research has been done in the domain of detecting Diabetic Retinopathy using deep learning [4,5].…”

Section: Introductionmentioning

confidence: 99%

DRNET: Improved Deep Learning Solution to detect Diabetic Retinopathy by incorporating Explainability, Uncertainty and Generalisation

Saxena

Narra

Saxena

2022

Preprint

View full text Add to dashboard Cite

Diabetic Retinopathy (DR) is the leading cause of blindness among adults and has no visible symptoms. Early detection is the key to prevent vision loss. Computer-Aided Diagnosis using Convolutional Neural Networks have recently gained momentum for DR detection as the cost can be significantly reduced while making the diagnosis more accessible. Even though these models give high accuracy, confidence in the diagnosis remains a mythical goal. In our opinion, the way to make the model’s diagnosis reliable is to enhance the trust in its predictions. In this work, we present a methodology to strengthen confidence in DR diagnosis using state of the art deep learning techniques by extending the pipeline with model uncertainty and explainability. Towards this goal, along with the model’s prediction we produce uncertainty scores and explainability maps to improve the users’ confidence in the diagnosis. Further, we attempt to generalise CNN models by exploring ensembling techniques such as bagging to improve performance of stand-alone models on external datasets i.e.-datasets other than the one that was used for training the model.

show abstract

Deep Learning Techniques for Diabetic Retinopathy Classification: A Survey

Cited by 123 publications

References 70 publications

Novel Framework for Enhanced Learning-based Classification of Lesion in Diabetic Retinopathy

Novel Framework for Enhanced Learning-based Classification of Lesion in Diabetic Retinopathy

MSF-Model: Multi-Scale Feature Fusion-Based Domain Adaptive Model for Breast Cancer Classification of Histopathology Images

DRNET: Improved Deep Learning Solution to detect Diabetic Retinopathy by incorporating Explainability, Uncertainty and Generalisation

Contact Info

Product

Resources

About