Data on OCT and fundus images for the detection of glaucoma

Raja, Hina; Akram, M. Usman; Khawaja, Sajid Gul; Arslan, Muhammad; Ramzan, Aneeqa; Nazir, Noman

doi:10.1016/j.dib.2020.105342

Cited by 22 publications

(15 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For a block of n consecutive convolution layers within the network having the dilation rate 'r' where n > r, the dilation factors in the proposed framework are generated through round(r − n 2 + i) where i varies from 0 to n − 1. For example, for a block containing 5 cascaded convolution layers having the dilation rate r = 3, the dilation factors will be [1,2,3,4,5], meaning that the first convolution layer within the block will perform ). Similarly, the green pixels (in layer 4) are computed through brown pixels in layer 3, brown pixels are computed through red pixels in layer 2 and red pixels are computed through dark blue pixels in layer 1.…”

Section: B Hybrid Convolution Frameworkmentioning

confidence: 99%

“…standard convolution (as r = 1), the second layer will have r = 2 and so on as shown in Figure 5 (bottom row). Similarly, for n = 3, r = 3, the dilation factors will be [2,3,4]. The second major benefit of RAG-Net v2 is that it is extremely lightweight and contains 91.04% fewer parameters than original RAG-Net architecture (having 62,352,188 parameters in total) while achieving the better segmentation and classification performance.…”

Section: B Hybrid Convolution Frameworkmentioning

confidence: 99%

“…Armed Forces Institute of Ophthalmology (AFIO) dataset, firstly introduced in [3], is a publicly available repository containing high-resolution ONH SD-OCT scans of healthy and glaucomatous subjects. The dataset has been acquired from AFIO Hospital, Rawalpindi, Pakistan.…”

Section: A Afio Datasetmentioning

confidence: 99%

“…We have also performed a series of clinical trials in which we cross-validated the predictions of the proposed framework with the recommendations from the expert clinicians. Table VIII reports samples of the clinical trials along with the recommendations of the four expert ophthalmologists (publicly available in the dataset package [3]). Furthermore, we report in Table IX the Pearson correlation analysis along with its statistical significance showcasing the clinical validation of the proposed framework with each clinician.…”

Section: F Clinical Trialsmentioning

confidence: 99%

See 3 more Smart Citations

Clinically Verified Hybrid Deep Learning System for Retinal Ganglion Cells Aware Grading of Glaucomatous Progression

Raja¹,

Hassan

Akram³

et al. 2021

IEEE Trans. Biomed. Eng.

Self Cite

View full text Add to dashboard Cite

Glaucoma is the second leading cause of blindness worldwide. Glaucomatous progression can be easily monitored by analyzing the degeneration of retinal ganglion cells (RGCs). Many researchers have screened glaucoma by measuring cup-to-disc ratios from fundus and optical coherence tomography scans. However, this paper presents a novel strategy that pays attention to the RGC atrophy for screening glaucomatous pathologies and grading their severity. Methods:The proposed framework encompasses a hybrid convolutional network that extracts the retinal nerve fiber layer, ganglion cell with the inner plexiform layer and ganglion cell complex regions, allowing thus a quantitative screening of glaucomatous subjects. Furthermore, the severity of glaucoma in screened cases is objectively graded by analyzing the thickness of these regions. Results: The proposed framework is rigorously tested on publicly available Armed Forces Institute of Ophthalmology (AFIO) dataset, where it achieved the F1 score of 0.9577 for diagnosing glaucoma, a mean dice coefficient score of 0.8697 for extracting the RGC regions and an accuracy of 0.9117 for grading glaucomatous progression. Furthermore, the performance of the proposed framework is clinically verified with the markings of four expert ophthalmologists, achieving a statistically significant Pearson correlation coefficient of 0.9236. Conclusion: An automated assessment of RGC degeneration yields better glaucomatous screening and grading as compared to the state-ofthe-art solutions. Significance: An RGC-aware system not only screens glaucoma but can also grade its severity and here we present an end-to-end solution that is thoroughly evaluated on a standardized dataset and is clinically validated for analyzing glaucomatous pathologies.

show abstract

Section: B Hybrid Convolution Frameworkmentioning

confidence: 99%

Section: B Hybrid Convolution Frameworkmentioning

confidence: 99%

Section: A Afio Datasetmentioning

confidence: 99%

Section: F Clinical Trialsmentioning

confidence: 99%

See 2 more Smart Citations

Clinically Verified Hybrid Deep Learning System for Retinal Ganglion Cells Aware Grading of Glaucomatous Progression

Raja¹,

Hassan

Akram³

et al. 2021

IEEE Trans. Biomed. Eng.

Self Cite

View full text Add to dashboard Cite

show abstract

“…At this point, it should be noted that a very recent work [62] also proposes a kind of glaucoma grading set up making use of the Armed Forces Institute of Ophthalmology (AFIO) data set [63], which contains OCT scans centred on the ONH. However, it pursues the discrimination between healthy, suspects and glaucomatous samples by computing the distance of different retinal layers of interest previously segmented.…”

Section: Contribution Of This Workmentioning

confidence: 99%