Retinal Cysts are formed by leakage and accumulation of fluid in the retina due to the incompetence of retinal vasculature. These cystic spaces have significance in several ocular diseases such as age-related macular degeneration, diabetic macular edema, etc. Optical coherence tomography is one of the predominant diagnosing techniques for imaging retinal pathologies. Segmenting and quantification of intraretinal cysts plays the vital role in predicting visual acuity. In literature, several methods have been proposed for automatic segmentation of intraretinal cysts. As cystoid macular edema becomes a major problem to humankind, we need to quantify it accurately and operate it out, else it might cause many problems later on. Though research is being carried out in this area, not much of progress has been made and accuracy achieved so far is 68% which is very less. Also, the methods depend on the quality of the image and give very low results for high noise images like topcon. This work uses ResNet CNN (Convolutional Neural Network) approach of segmentation by the way of patchwise classification for training on image set from cyst segmentation challenge dataset and testing on test data set given by 2 different graders for all 4 vendors in the challenge. It also compares these methods using first publicly available novel cyst segmentation challenge dataset. The methods were evaluated using quantitative measures to assess their robustness against the challenges of intraretinal cyst segmentation. The results are found to be better than the previous state of the art approaches giving more than 70% dice coefficient on all vendors irrespective of their quality.
Figures & tables
Fig. 1: Figure(above) represents 7 intra retinal layers with cyst in it. Figure(below) shows true cystic regions in its ground truth image.
Fig. 2: Feedforward network
Fig. 3: Feedback network.
Fig. 4: Residual Network information flow.
Fig. 5: Feedforward Network vs Residual Network
Set
Cirrus
Spectralis
Topcon
Nidek
Total
Training
4
4
4
4
15
Testing 1
2
2
2
2
8
Testing 2
2
2
2
1
7
TABLE I: Description of Dataset
Fig. 6: Two actual classes or observed labels
Fig. 7: Predicted classes of a perfect classifier
Fig. 8: Predicted classes of a classifier
Fig. 9: Four outcomes of a classifier
Volume
Dice Coefficient
Sensitivity/Recall
Precision
Cirrus_1
0.8893
0.8018
0.9982
Cirrus_2
0.8156
0.6892
0.9986
Cirrus_3
0.917
0.8475
0.999
Cirrus_4
0.93
0.8721
0.9961
Mean
0.888
0.8027
0.998
Std
0.0511
0.0810
0.0013
TABLE II: Test Results for Grader 1
Volume
Dice Coefficient
Sensitivity/Recall
Precision
Cirrus 1
0.8893
0.8019
0.9981
Cirrus_2
0.8155
0.6892
0.9986
Cirrus_3
0.9171
0.8476
0.999
Cirrus_4
0.9302
0.8726
0.996
Mean
0.8880
0.8028
0.9979
Std
0.0513
0.0812
0.0013
TABLE III: Test Results for Grader 2
Volume
Dice Coefficient
Sensitivity/Recall
Precision
Cirrus 1
0.8893
0.802
0.9979
Cirrus_2
0.8154
0.6892
0.9981
Cirrus_3
0.9171
0.8477
0.9988
Cirrus_4
0.9305
0.8733
0.9957
Mean
0.8880
0.8031
0.9976
Std
0.0514
0.0814
0.0013
TABLE IV: Test Results for Grader 3 (Intersection of Grader 1 & 2)
Team
Mean Dice Coeff
Std. Deviation
de Sisternes et al
0.68
0.14
Venhuizen et al.
0.601
0.18
Oguz et al.
0.596
0.14
Esmaeili et al.
0.55
0.24
Haritz et al.
0.23
0.15
Proposed approach
0.82
0.08
TABLE V: Results of various State of art methods compared with the Proposed Approach
Measuring retinal fluid from optical coherence tomography (OCT) drives treatment decisions in macular disease, but manual annotation is slow and segmentation models trained on one scanner degrade on another. We present an attention-guided TransUNet that segments three fluid types across four independent OCT sources, combining a domain-adaptive normalisation scheme with an uncertainty estimate that flags unreliable pixels. The model reaches a mean fluid Dice of 0.78, and -- most usefully for clinicians -- its uncertainty is 1.34x higher exactly where expert graders disagree (p<10^-4), turning a raw segmentation map into an actionable clinical triage signal.
Animesh Kumar
Newcastle University, Newcastle upon Tyne NE1 7RU, United Kingdom
Reliable automated analysis of Optical Coherence Tomography (OCT) imaging is crucial for diagnosing retinal disorders but faces a critical barrier: the need for expensive, labor-intensive expert annotations. Supervised deep learning models struggle to generalize across diverse pathologies, imaging devices, and patient populations due to their restricted vocabulary of annotated abnormalities. We propose an unsupervised anomaly detection framework that learns the normative distribution of healthy retinal anatomy without lesion annotations, directly addressing annotation efficiency challenges in clinical deployment. Our approach leverages a discrete latent model trained on normal B-scans to capture OCT-specific structural patterns. To enhance clinical robustness, we incorporate retinal layer-aware supervision and structured triplet learning to separate healthy from pathological representations, improving model reliability across varied imaging conditions. During inference, anomalies are detected and localized via reconstruction discrepancies, enabling both image and pixel-level identification without requiring disease-specific labels. On the Kermany dataset (AUROC: 0.799), our method substantially outperforms VAE, VQVAE, VQGAN, and f-AnoGAN baselines. Critically, cross-dataset evaluation on Srinivasan achieves AUROC 0.884 with superior generalization, demonstrating robust domain adaptation. On the external RETOUCH benchmark, unsupervised anomaly segmentation achieves competitive Dice (0.200) and mIoU (0.117) scores, validating reproducibility across institutions.
Tania Haghighi, Sina Gholami, Hamed Tabkhi +1
University of North Carolina at Charlotte Charlotte, NC, USA
Early and accurate classification of retinal diseases is critical to counter vision loss and for guiding clinical management of retinal diseases. In this study, we proposed a deep learning method for retinal disease classification utilizing optical coherence tomography (OCT) images from the Retinal OCT Image Classification - C8 dataset (comprising 24,000 labeled images spanning eight conditions). Images were resized to 224x224 px and tested on convolutional neural network (CNN) architectures: Xception and InceptionV3. Data augmentation techniques (CutMix, MixUp) were employed to enhance model generalization. Additionally, we applied GradCAM and LIME for interpretability evaluation. We implemented this in a real-world scenario via our web application named RetinaVision. This study found that Xception was the most accurate network (95.25%), followed closely by InceptionV3 (94.82%). These results suggest that deep learning methods allow effective OCT retinal disease classification and highlight the importance of implementing accuracy and interpretability for clinical applications.
Mohammad Tahmid Noor, Shayan Abrar, Jannatul Adan Mahi +3
Dept. of CSE East West University Dhaka, Bangladesh · Dept. of CSE American International University-Bangladesh Dhaka, Bangladesh · Dept. of ICT Bangladesh Army University of Engineering and Technology Dhaka, Bangladesh