Graph-Cut Segmentation of Retinal Layers from OCT Images

Dodo, Bashir Isa; Li, Yongmin; Eltayef, Khalid; Liu, Xiaohui

doi:10.5220/0006580600350042

Cited by 14 publications

(8 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…5. Although the methods in [17], [18] have no outliers beyond the first quartile, their accuracy values start from ∼ 76% which is low compared to the starting value of ∼ 84% of the proposed method. The method in [20] has the second-best start of the first quartile.…”

Section: Resultsmentioning

confidence: 74%

“…Moreover, the methods in [17], [42] segmented the RPE regions better than the proposed method. Their performance is due to the region limitation, whereby the methods can search within a specific range based on the layer characteristics, and the RPE region has limited blood vessels interference.…”

Section: Resultsmentioning

confidence: 89%

“…Inherently, due to the regional competition in [18], it adapts to the RPE region topology better than the proposed method, although both methods do not employ region limitation or topology constraints. Generally, region limitation based on retinal layer topology aids in handling incompleteness in [17], [20], [42], while FCM aids in handling inhomogeneity within layers by adaptively estimating values specific to each image in [18] and the proposed method. This can be deduced from the NFL layer performance discussed earlier based on the performance matrix in Table 1, and in comparison of [18] and the proposed method to [17], [42] for the level set and graph cut methods, 2.…”

Section: Resultsmentioning

confidence: 99%

“…The segmentation is achieved by minimising a cost function which adds the values of the weights in the edges that are separated. In [17], Dodo et al proposed an automated graph-cut segmentation method of retinal layers from OCT scans. This combines a fuzzy histogram hyperbolisation model and graph cut algorithm to segment eight (8) intraretinal layers in high-resolution 3D SD-OCT images.…”

Section: Related Workmentioning

confidence: 99%

“…the boundary between the retinal nerve fibre layer and the vitreous, and the boundary between the RPE and the choroid regions. Additionally, it is commonly accepted that the NFL, IS-OS and RPE exhibit high reflectivity in an OCT image [42]- [44], and based on experiments the ILM and RPE exhibits the highest transitions from dark-bright and bright-dark, respectively [17]. With these understanding of the retinal structure, we simply identify the ILM and RPE using the shortest path [45], by searching for the highest transitions on two separate adjacency matrices [42].…”

Section: A Preprocessingmentioning

confidence: 99%

See 4 more Smart Citations

Retinal Layer Segmentation in Optical Coherence Tomography Images

Dodo

Kaba

et al. 2019

IEEE Access

Self Cite

View full text Add to dashboard Cite

The four major causes of blindness are age-related diseases, out of which three affects the retina. These diseases, i.e., glaucoma, diabetic retinopathy, and age-related macular degeneration, require life-long treatment and cause irreversible blindness. Conversely, early diagnosis has been shown to curtail or prevent blindness and visual impairments. A critical element of the clinical diagnosis is the analysis of individual retinal layer properties, as the manifestation of the dominant eye diseases has been shown to correlate with structural changes to the retinal layers. Regrettably, manual segmentation is dependent on the ophthalmologist's level of expertise, and currently becoming impractical due to advancement in imaging modalities. Inherently, much research on computer-aided diagnostic methods is conducted to aid in extracting useful layer information from these images, which were inaccessible without these techniques. However, speckle noise and intensity inhomogeneity remain a challenge with a detrimental effect on the performance of automated methods. In this paper, we propose a method comprising of fuzzy image processing techniques and graph-cut methods to robustly segment optical coherence tomography (OCT) into five (5) distinct layers. Notably, the method establishes a specific region of interest to suppress the interference of speckle noise, while Fuzzy C-means is utilized to build data terms for better integration into the continuous max-flow to handle inhomogeneity. The method is evaluated on 225 OCT B-scan images, and promising experimental results were achieved. The method will allow for early diagnosis of major eye diseases by providing the basic, yet critical layer information necessary for an effective eye examination.

show abstract

Section: Resultsmentioning

confidence: 74%

Section: Resultsmentioning

confidence: 89%