Incremental learning for exudate and hemorrhage segmentation on fundus images

He, Wanji; Wang, Xin; Wang, Lin; Huang, Yelin; Yang, Zhiwen; Yao, Xuan; Zhao, Xin; Ju, Lie; Wu, Liao; Wu, Lin; Lu, Huimin; Ge, Zongyuan

doi:10.1016/j.inffus.2021.02.017

Cited by 27 publications

(5 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…34,35 added the residual structure to the network to achieve segmentation. 36 used incremental learning to avoid retrospective interference caused by learning new features. 37 proposed the use of a support vector machine classifier and a fast R-CNN to preliminarily filtre the input images, and discard the images without hard exudative lesions, leaving only the images with hard exudative lesions.…”

Section: Discussionmentioning

confidence: 99%

Nonproliferative diabetic retinopathy dataset(NDRD): A database for diabetic retinopathy screening research and deep learning evaluation

Liang,

Wen,

Duan

et al. 2024

Health Informatics J

View full text Add to dashboard Cite

Objectives: In this article, we provide a database of nonproliferative diabetes retinopathy, which focuses on early diabetes retinopathy with hard exudation, and further explore its clinical application in disease recognition. Methods: We collect the photos of nonproliferative diabetes retinopathy taken by Optos Panoramic 200 laser scanning ophthalmoscope, filter out the pictures with poor quality, and label the hard exudative lesions in the images under the guidance of professional medical personnel. To validate the effectiveness of the datasets, five deep learning models are used to perform learning predictions on the datasets. Furthermore, we evaluate the performance of the model using evaluation metrics. Results: Nonproliferative diabetes retinopathy is smaller than proliferative retinopathy and more difficult to identify. The existing segmentation models have poor lesion segmentation performance, while the intersection over union ( IOU) value for deep lesion segmentation of models targeting small lesions can reach 66.12%, which is higher than ordinary lesion segmentation models, but there is still a lot of room for improvement. Conclusion: The segmentation of small hard exudative lesions is more challenging than that of large hard exudative lesions. More targeted datasets are needed for model training. Compared with the previous diabetes retina datasets, the NDRD dataset pays more attention to micro lesions.

show abstract

Section: Discussionmentioning

confidence: 99%

Nonproliferative diabetic retinopathy dataset(NDRD): A database for diabetic retinopathy screening research and deep learning evaluation

Liang,

Wen,

Duan

et al. 2024

Health Informatics J

View full text Add to dashboard Cite

show abstract

“…These include their computational complexity, the requirement for large datasets, their dependence on data quality, and their sensitivity to variable initialization. An incremental learning strategy was proposed in [15], which utilizes both historical and contemporary segmentation models to improve segmentation results.…”

Section: Exudates Segmentationmentioning

confidence: 99%

Hybrid Classification Approach Utilizing DenseUNet+ for Diabetic Macular Edema Disorder Detection

Abualigah,

Almomani,

Abu Zitar

et al. 2024

Int. J. Onl. Eng.

View full text Add to dashboard Cite

Diabetic macular edema (DME) poses a significant threat to vision. It is characterized by the enlargement of the macula due to the accumulation of plasma in the extracellular space of the retina. Detection of DME, crucial for timely intervention, traditionally relies on manual inspection of images, which is time-consuming and prone to human error. Leveraging advancements in computer-assisted diagnostics, this study proposes a novel approach utilizing the DenseUNet+ architecture tailored for precise segmentation across diverse image modalities. The proposed method integrates data from four modalities within dense block structures, followed by linear operations and concatenation to enhance feature representation. Evaluation using ResNet101V2 and DenseNet201 demonstrates superior performance, with accuracy exceeding 95% and 99%, respectively, showcasing their efficacy in screening retinal optical coherence tomography (OCT) images for DME. This research highlights the potential of deep learning techniques to enhance ophthalmologists’ abilities to efficiently screen, diagnose, and manage DME, thus reducing the risk of irreversible vision loss.

show abstract

“…Many DL detection/segmentation methods focusing on pixel-by-pixel annotation often lead to a danger of catastrophic interference, where the model abruptly forgets the previously learned attributes while learning new information. He et al [75] in their work used incremental learning to avoid this problem, where the knowledge of the previous model is utilized to perfect the present model.…”

Section: ) Dr Diagnosis Using Exudatesmentioning

confidence: 99%

A Comprehensive Review of Deep Learning Strategies in Retinal Disease Diagnosis Using Fundus Images

et al. 2022

View full text Add to dashboard Cite

In recent years, there has been an unprecedented growth in computer vision and deep learning implementation owing to the exponential rise of computation infrastructure. The same was also reflected in retinal image analysis and successful artificial intelligence models were developed for various retinal disease diagnoses using a wide variety of visual markers obtained from eye fundus images. This article presents a comprehensive study of different deep learning strategies employed in recent times for the diagnosis of five major eye diseases, i.e., Diabetic retinopathy, Glaucoma, age-related macular degeneration, Cataract, and Retinopathy of prematurity. This article is organized according to the deep learning implementation process pipeline, where commonly used datasets, evaluation metrics, image pre-processing techniques, and deep learning backbone models are first illustrated followed by an extensive review of different strategies for each of the five mentioned retinal diseases is presented. Finally, this article summarizes eight major research directions available in the field of retinal disease diagnosis and outlines key challenges and future scope for the present research community.

show abstract

Incremental learning for exudate and hemorrhage segmentation on fundus images

Cited by 27 publications

References 13 publications

Nonproliferative diabetic retinopathy dataset(NDRD): A database for diabetic retinopathy screening research and deep learning evaluation

Nonproliferative diabetic retinopathy dataset(NDRD): A database for diabetic retinopathy screening research and deep learning evaluation

Hybrid Classification Approach Utilizing DenseUNet+ for Diabetic Macular Edema Disorder Detection

A Comprehensive Review of Deep Learning Strategies in Retinal Disease Diagnosis Using Fundus Images

Contact Info

Product

Resources

About