Macula segmentation and fovea localization employing image processing and heuristic based clustering for automated retinal screening

GeethaRamani, R.; Balasubramanian, Lakshmi

doi:10.1016/j.cmpb.2018.03.020

Cited by 28 publications

(5 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some attempts have been made to localize the macula without relying on anatomical features. For example, GeethaRamani and Balasubramanian [11] propose an approach to segment the macula using an unsupervised clustering algorithm. Pachade et al [13] directly select the square in the middle of the image as ROI and use a filter on intensity for fovea localization.…”

Section: A Anatomical Structure-based Methodsmentioning

confidence: 99%

“…Bilateral-ViT [15] is the only previous deep learning-based method that incorporates fundus and vessel features. However, in most deep learningbased studies [11], [13], [14], [22]- [25], only fundus images are used, resulting in poor incorporation of anatomical relationships throughout the entire image, leading to failure in more challenging cases.…”

Section: A Comparison To State Of the Artmentioning

confidence: 99%

“…Fovea localization is an important upstream task in clinical diagnosis, helping to diagnose maculopathy and abnormalities [2], [3]. Although recent works have employed deep learning methods to improve performance, they typically only utilize fundus images as input [11], [13], [14], [21]- [25]. These works are also implemented on datasets containing few challenging images, resulting in three main pitfalls: 1) inadequate exploitation of the anatomical structure outside the macula as only fundus images are used as input; 2) typical convolution-based architectures lacking incorporation of global features; and 3) sensitivity to challenging cases, such as rare fovea positions and severe lesions.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Bilateral-Fuser: A Novel Multi-cue Fusion Architecture with Anatomical-aware Tokens for Fovea Localization

Song¹,

Wang²,

Wang³

et al. 2023

Preprint

View full text Add to dashboard Cite

Accurate localization of the fovea is a crucial initial step in analyzing retinal diseases since it helps prevent irreversible vision loss. Although current deep learning-based methods achieve better performance than traditional methods, they still face challenges such as inadequate utilization of anatomical landmarks, sensitivity to diseased retinal images, and various image conditions. In this paper, we propose a novel transformer-based architecture (Bilateral-Fuser) for multi-cue fusion. The Bilateral-Fuser explicitly incorporates long-range connections and global features using retina and vessel distributions to achieve robust fovea localization. We introduce a spatial attention mechanism in the dual-stream encoder to extract and fuse self-learned anatomical information. This design focuses more on features distributed along blood vessels and significantly reduces computational costs by reducing token numbers. Our comprehensive experiments demonstrate that the proposed architecture achieves state-of-theart performance on two public datasets and one large-scale private dataset. Moreover, we show that the Bilateral-Fuser is more robust on both normal and diseased retina images and has better generalization capacity in cross-dataset experiments.

show abstract

Section: A Anatomical Structure-based Methodsmentioning

confidence: 99%

Section: A Comparison To State Of the Artmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Bilateral-Fuser: A Novel Multi-cue Fusion Architecture with Anatomical-aware Tokens for Fovea Localization

Song¹,

Wang²,

Wang³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Then, a classification based on the "Gaussian mixture model" (GMM) algorithm is applied to classify candidate regions as macular or non-macular. Similarly, intensity is used to segment the macula and identify the fovea in the method proposed by [19] et al [20] and Kumar et al [21]. The intensity feature is combined with the fovea -OD distance in the methods proposed by Kamble et al [22], Singh et al [23] and Rahim et al [24].…”

Section: Related Work: Synthesis Of Macular Location Methodsmentioning

confidence: 99%

A Robustness Study of Machine Learning Based Methods for Macula Detection in Pathological Fundus Images

Elloumi,

Kachouri

2023

2023 Twelfth International Conference on Image Processing Theory, Tools and Applications (IPTA)

View full text Add to dashboard Cite

The macula detection is an important step for screening retinal diseases. Several methods have been proposed for detecting macula, with the aim of achieving a higher detection performance despite retinal pathologies. Their processing principles are based on well-known anatomical criteria of macula. However, these criteria are subject to significant changes caused by various eye pathologies. Therefore, it leads to a non-performing macula detection, and hence pathology screening. The objective of this work is to study the robustness of the macula detection methods in terms of retinal pathologies. For this purpose, the main macular criteria are extracted. Then, recent macula detection methods are detailed where the macula criteria taken into account are identified. Thereafter, the retinal pathologies that affect the macula are listed where caused macula changes are figured out. This study allows us to synthesise the robustness of each method in terms of pathologies.

show abstract

“…Clustering is a frequently-used data mining tool for unlabeled data analyzing and a significant branch of unsupervised machine learning, the distribution of sample data could be obtained by clustering [1,2]. Clustering has been applied in many real-world applications: image processing [3][4][5], text organization [6], food detection [7], bioinformatics [8], etc.…”

Section: Introductionmentioning

confidence: 99%

Adaptive local neighborhood information based efficient fuzzy clustering approach

Wu,

Zhao,

et al. 2024

Complex Intell. Syst.

View full text Add to dashboard Cite

The purpose of clustering is to partition data similar with each other into a same group and partition data dissimilar with each other into different groups. However, in most existing fuzzy clustering approaches, the membership degrees of an individual belonging to different clusters are relied on the different distances between the individual and different cluster centroids, the similarity between the individual and data in different clusters are ignored, besides, the outliers cannot be distinguished effectively. For improving the clustering performance and addressing the problems in most existing fuzzy clustering approaches, based on the concept that data close to each other should be grouped together and be assigned to a same cluster, in this paper, we present a novel efficient fuzzy clustering approach, in which the adaptive local neighborhood information of each data referring to different clusters is taken into consideration; an entirely new idea that the membership degree values of an individual referring to different clusters should not only depend on the distances between the individual and different cluster centers, but also rely on the distances between the individual and several nearest neighbor data in different clusters is put forward; a new scheme to search for the outliers is presented, a method for identifying the different importance of different features is introduced. Experiments on synthetic and publicly real-world datasets were conducted, the clustering results show that the approach put forward can yield consistently satisfying clustering performance and has significant advantages.

show abstract

Macula segmentation and fovea localization employing image processing and heuristic based clustering for automated retinal screening

Cited by 28 publications

References 14 publications

Bilateral-Fuser: A Novel Multi-cue Fusion Architecture with Anatomical-aware Tokens for Fovea Localization

Bilateral-Fuser: A Novel Multi-cue Fusion Architecture with Anatomical-aware Tokens for Fovea Localization

A Robustness Study of Machine Learning Based Methods for Macula Detection in Pathological Fundus Images

Adaptive local neighborhood information based efficient fuzzy clustering approach

Contact Info

Product

Resources

About