Comment on: 'AI deepfake: GPT-4o can produce near-authentic fundus images'.

Age-related macular degeneration (ARMD) is one of the most common retinal syndromes that occurs in elderly people. Different eye testing techniques such as fundus photography and optical coherence tomography (OCT) are used to clinically examine the ARMD-affected patients. Many researchers have worked on detecting ARMD from fundus images, few of them also worked on detecting ARMD from OCT images. However, there are only few systems that establish the correspondence between fundus and OCT images to give an accurate prediction of ARMD pathology. In this paper, we present fully automated decision support system that can automatically detect ARMD by establishing correspondence between OCT and fundus imagery. The proposed system also distinguishes between early, suspect and confirmed ARMD by correlating OCT B-scans with respective region of the fundus image. In first phase, proposed system uses different B-scan based features along with support vector machine (SVM) to detect the presence of drusens and classify it as ARMD or normal case. In case input OCT scan is classified as ARMD, region of interest from corresponding fundus image is considered for further evaluation. The analysis of fundus image is performed using contrast enhancement and adaptive thresholding to detect possible drusens from fundus image and proposed system finally classified it as early stage ARMD or advance stage ARMD. The proposed system is tested on local data set of 100 patients with100 fundus images and 6800 OCT B-scans. Proposed system detects ARMD with the accuracy, sensitivity, and specificity ratings of 98.0, 100, and 97.14%, respectively.

A hybrid segmentation algorithm is proposed is this paper to extract the blood vessels from the fundus image of retina. Fundus camera captures the posterior surface of the eye and the captured images are used to diagnose diseases, like Diabetic Retinopathy, Retinoblastoma, Retinal haemorrhage, etc. Segmentation or extraction of blood vessels is highly required, since the analysis of vessels is crucial for diagnosis, treatment planning, and execution of clinical outcomes in the field of ophthalmology. It is derived from the literature review that no unique segmentation algorithm is suitable for images of different eye-related diseases and the degradation of the vessels differ from patient to patient. If the blood vessels are extracted from the fundus images, it will make the diagnosis process easier. Hence, this paper aims to frame a hybrid segmentation algorithm exclusively for the extraction of blood vessels from the fundus image. The proposed algorithm is hybridized with morphological operations, bottom hat transform, multi-scale vessel enhancement (MSVE) algorithm, and image fusion. After execution of the proposed segmentation algorithm, the area-based morphological operator is applied to highlight the blood vessels. To validate the proposed algorithm, the results are compared with the ground truth of the High-Resolution Fundus (HRF) images dataset. Upon comparison, it is inferred that the proposed algorithm segments the blood vessels with more accuracy than the existing algorithms.

BackgroundWith the diabetes mellitus (DM) prevalence increasing annually, the human grading of retinal images to evaluate DR has posed a substantial burden worldwide. SmartEye is a recently developed fundus image processing and analysis system with lesion quantification function for DR screening. It is sensitive to the lesion area and can automatically identify the lesion position and size. We reported the diabetic retinopathy (DR) grading results of SmartEye versus ophthalmologists in analyzing images captured with non-mydriatic fundus cameras in community healthcare centers, as well as DR lesion quantitative analysis results on different disease stages.MethodsThis is a cross-sectional study. All the fundus images were collected from the Shanghai Diabetic Eye Study in Diabetics (SDES) program from Apr 2016 to Aug 2017. 19,904 fundus images were acquired from 6013 diabetic patients. The grading results of ophthalmologists and SmartEye are compared. Lesion quantification of several images at different DR stages is also presented.ResultsThe sensitivity for diagnosing no DR, mild NPDR (non-proliferative diabetic retinopathy), moderate NPDR, severe NPDR, PDR (proliferative diabetic retinopathy) are 86.19, 83.18, 88.64, 89.59, and 85.02%. The specificity are 63.07, 70.96, 64.16, 70.38, and 74.79%, respectively. The AUC are PDR, 0.80 (0.79, 0.81); severe NPDR, 0.80 (0.79, 0.80); moderate NPDR, 0.77 (0.76, 0.77); and mild NPDR, 0.78 (0.77, 0.79). Lesion quantification results showed that the total hemorrhage area, maximum hemorrhage area, total exudation area, and maximum exudation area increase with DR severity.ConclusionsSmartEye has a high diagnostic accuracy in DR screening program using non-mydriatic fundus cameras. SmartEye quantitative analysis may be an innovative and promising method of DR diagnosis and grading.

The paper proposes a novel method for extraction of blood vessels and veins from medical image of human eye - retinal fundus images that can be used in ophthalmology for detecting various eyes' diseases such glaucoma, diabetic retinopathy or macula oedema. The method utilizes an approach of preprocessing of image by using adaptive histogram equalization by CLAHE algorithm of green channel of fundus retinal image. Subsequently, using adaptive filters and image convolution with filter mask as key point of proposed algorithm and subsequently is applied the operation erosion processed image and removed small segments from image to enhance extraction of blood vessels from fundus image. The proposed technique analyzes detection and evaluates precision of the method on dataset from public fundus image libraries DRIVE, and HRF and compare with reference training results provided by these libraries..

We previously developed a technique to acquire a SLO (scanning laser ophthalmoscope) like fundus intensity image from the raw spectra measured with spectral-domain optical coherence tomography (OCT), the same spectra used to generate a 3D OCT data set. This technique offers simultaneous fundus and OCT images and, therefore, solves the problem of registering a cross sectional OCT image to fundus features. However, the registration of high density OCT images is still an unsolved problem because no useful fundus image can be generated from the high density scans. High density OCT images can significantly improve the image quality and enhance the visualization of retinal structure, especially the structure of small lesions. We have developed a feature-based algorithm, which can register a high density OCT image on the fundus image generated from normal density scans. The algorithm was successfully tested for both normal and diseased eyes.

Many people lose sight due to diabetic retinopathy. The reason that diabetic retinopathy is dangerous is that it cannot return to its pre-onset state after the disease's onset. Most patients take fundus images that capture the retina, and the doctor uses the fundus images to determine the presence of disease. Existing fundus images could only identify a narrow range, making it difficult to diagnose the disease accurately. However, with technological advances, ultra-wide-field fundus images that allow the wider retina to be seen have emerged. However, in deep learning research, many studies use existing fundus images due to the lack of new data. In the case of new technologies such as ultra-wide-field fundus images, it was often difficult to obtain data, so deep learning research could not be done properly. In the case of ultra-wide-field fundus images, research was conducted using data from hundreds to ten thousand sheets, but compared to large-scale data sets, the deep learning performance is inevitably inferior compared to large-scale data sets. In this study, synthetic data were created using ultra-wide-field fundus images and various GAN models to solve this problem. As a result of the study, BEGAN was derived similarly to the real image in qualitative and quantitative evaluation. However, it fell into mode collapse and showed the same output even when a new input came in. Mode collapse in BEGAN could be appeared depending on the amount and size of data, so various studies using BEGAN are needed.

Anomaly detection in fundus images remains challenging due to the fact that fundus images often contain diverse types of lesions with various properties in locations, sizes, shapes, and colors. Current methods achieve anomaly detection mainly through reconstructing or separating the fundus image background from a fundus image under the guidance of a set of normal fundus images. The reconstruction methods, however, ignore the constraint from lesions. The separation methods primarily model the diverse lesions with pixel-based independent and identical distributed (i.i.d.) properties, neglecting the individualized variations of different types of lesions and their structural properties. And hence, these methods may have difficulty to well distinguish lesions from fundus image backgrounds especially with the normal personalized variations (NPV). To address these challenges, we propose a patch-based non-i.i.d. mixture of Gaussian (MoG) to model diverse lesions for adapting to their statistical distribution variations in different fundus images and their patch-like structural properties. Further, we particularly introduce the weighted Schatten p-norm as the metric of low-rank decomposition for enhancing the accuracy of the learned fundus image backgrounds and reducing false-positives caused by NPV. With the individualized modeling of the diverse lesions and the background learning, fundus image backgrounds and NPV are finely learned and subsequently distinguished from diverse lesions, to ultimately improve the anomaly detection. The proposed method is evaluated on two real-world databases and one artificial database, outperforming the state-of-the-art methods.

Rapid development in the field of ophthalmology has increased the demand of computer aided diagnosis of various eye diseases. Papilledema is an eye disease in which the optic disc of the eye is swelled due to an increase in intracranial pressure. This increased pressure can cause severe encephalic complications like abscess, tumors, meningitis or encephalitis, which may lead to a patient's death. Although there have been several papilledema case studies reported from a medical point of view, only a few researchers have presented automated algorithms for this problem. This paper presents a novel computer aided system which aims to automatically detect papilledema from fundus images. Firstly, the fundus images are preprocessed by going through optic disc detection and vessel segmentation. After preprocessing, a total of 26 different features are extracted to capture possible changes in the optic disc due to papilledema. These features are further divided into four categories based upon their color, textural, vascular and disc margin obscuration properties. The best features are then selected and combined to form a feature matrix that is used to distinguish between normal images and images with papilledema using the supervised support vector machine (SVM) classifier. The proposed method is tested on 160 fundus images obtained from two different data sets i.e. structured analysis of retina (STARE), which is a publicly available data set, and our local data set that has been acquired from the Armed Forces Institute of Ophthalmology (AFIO). The STARE data set contained 90 and our local data set contained 70 fundus images respectively. These annotations have been performed with the help of two ophthalmologists. We report detection accuracies of 95.6% for STARE, 87.4% for the local data set, and 85.9% for the combined STARE and local data sets. The proposed system is fast and robust in detecting papilledema from fundus images with promising results. This will aid physicians in clinical assessment of fundus images. It will not take away the role of physicians, but will rather help them in the time consuming process of screening fundus images.

Screening of eye is essential to identify a range of retinal irregularities, and Fundus Imaging (FI) is one of the techniques commonly used by the ophthalmologist to record and evaluate the retinal abnormalities. Manual appraisal of FI is time-consuming, and hence, a computer-based methodology is always preferred. This work aims to develop a system to examine the retinal abnormality with the help of FI pictures. This work employs a hybrid processing scheme to enhance and extract the FI in order to extract and assess the optic-disc section. In this work, enhancement is done with firefly algorithm and Kapur’s thresholding, and the extraction is implemented with level set technique. All these testing are put into practice with MATLAB software using the benchmark Rim-One FI database, and the results of this study confirmed that the anticipated technique offered enhanced outcome in extracting the disc section from the RGB scaled FI.

This study presents an innovative Ensemble Disease Learning Algorithm (EDL) for the detection and classification of retinal diseases using fundus images. We enhance our method by incorporating deep learning techniques and multi-modal imaging data, including optical coherence tomography (OCT) images alongside fundus photographs, to provide a more comprehensive understanding of retinal pathology. The advanced EDL integrates Convolutional Neural Networks (CNNs) and attention mechanisms with Capsule Networks (CapsNet) and Support Vector Machine (SVM) classifiers for more nuanced feature extraction and classification. We introduce a novel ensemble adaptive weighting approach that dynamically adjusts classifier weights based on performance across disease types and severity levels, significantly improving the algorithm's handling of complex and rare cases. To enhance model interpretability, we implement an explainable AI component that provides visual heatmaps of the most significant regions for each diagnosis to clinicians. We evaluate the enhanced EDL on a large, diverse dataset encompassing multiple retinal diseases, including diabetic retinopathy, age-related macular degeneration, and glaucoma, across various ethnicities and age groups. Our results demonstrate superior accuracy, sensitivity, and specificity compared to our previous model and other state-of-the-art approaches. A prospective clinical validation study assesses the algorithm's real-world performance. This research advances automated retinal disease diagnosis by making it more robust, accurate, and clinically relevant, potentially improving patient outcomes and global eye care through early disease detection and treatment planning.

INTRODUCTION: Diabetic retinopathy (DR) is one of the consequences of diabetes which if untreated may lead to loss of vision. Generally, for DR detection, retinal images are obtained using a traditional fundus camera. A recent trend in the acquisition of eye fundus images is the usage of smartphones to acquire images.
 OBJECTIVES: This paper focuses on the study of existing works which incorporated smartphones for obtaining fundus images and various devices available in the market. Also, the common datasets used for carrying out DR detection using smartphone-based fundus images as well as the classification models used for the diagnosis of DR are explored.
 METHODS: A search of information was carried out on articles based on DR detection from fundus images published in the state-of-the-art literatures.
 RESULTS: Majority of the works uses SBFI devices like 20D lens, EyeExaminer etc. to obtain fundus image. The common databases used for the study are EyePACS, Messidor, etc. and the classification models mostly rely on deep learning frameworks.
 CONCLUSION: The use of smartphones for capturing fundus images for DR detection are explored. Smartphone devices, datasets used for the study and currently available classification models for SBFI based DR detection are discussed in detail. This paper portrays various approaches currently being employed in SBFI based DR detection.

Cataracts are a leading cause of blindness in Pakistan, contributing to more than 54% of cases due to poor living condition, nutritional deficiencies, and limited healthcare access. Early detection is critical to avoid invasive treatments,but current diagnostic approaches often identify cataracts at advanced stages. This paper presents an advanced,automated cataract detection system using deep learning specifically the ResNet-50 architecture, to address this gap. The model processes fundus retinal images curated from diverse datasets, classified by ophthalmologic experts through a rigorous three-stage process. By leveraging the ResNet-50 model, cataracts are categorized into normal,moderate,and severe, achieving an accuracy of 97.56% on full images. Notably, the system performs well even on partial images with 70% visibility, maintaining an accuracy of 95.23%, thus minimizing the need for extensive images restoration. The dataset was augmented to include 17,500 images,ensuring robust training. The model's ability to detect cataracts with high precision in images with varying visibility(70% ,80%,85% and beyond) demonstrate its flexibility and reliability, consistently achieving accuracy above 95.50%. This research offers a non-invasive, efficient solution particularly suited for remote areas, addressing the limitations of the late-stage diagnoses. It represent a significant advancement in cataract detection and has the potential to revolutionize global cataracts identification through early, accurate intervention.

ABSTRACTThe early detection of diabetic retinopathy (DR) illnesses improves diagnosis and lowers the risk of permanent blindness. As a result, screening for DR in fundus images is an important method used to diagnose diabetes and other eye diseases. However, detecting diseases manually requires a significant amount of time and work. Deep learning (DL) techniques have produced encouraging results in categorizing fundus images. Still, the multi‐class DR disease remains a difficult task. Thus, the proposed framework adopted a novel Criss‐Cross Attention‐Based Squeeze‐and‐Excitation Assisted Vision Transformer (CCA‐SE‐ViT) model to classify DR from fundus images. Initially, the defective region of the retina is segmented using a novel dilated depth‐wise separable convolutional U‐Net model (dDSC‐UNet). Then, using the segmented regions, the fundus images are classified into multiple classes as age‐related macular degeneration (AMD), DR, glaucoma, cataracts, myopia, hypertension, normal, other abnormalities, and DR cases are classified into no DR, mild, moderate, severe, and proliferative DR, respectively. The retinal fundus images are obtained from publicly available datasets like OIA‐ODIR and APTOS 2019. The proposed methodology for multi‐class categorization of retinal illnesses in the OIA‐ODIR dataset yielded 97.2% accuracy, 96.7% precision, 96.1% recall, 95.9% F1‐score, and 96.4% specificity. The APTOS dataset was used for multi‐class classification of DR illnesses, and the results were 99.68% accuracy, 99.08% precision, 99.31% recall, 99.19% F1‐score, and 99.26% specificity. The results demonstrated that the proposed method accurately identifies DR using retinal fundus images.

In this paper, we present the previous development and deployment of Fundus Analysis Software Tool (FAST) to enable the analysis of different anatomical features and pathologies within fundus images over time, and demonstrate its usefulness with three use cases. First, we utilized FAST to acquire 616 fundus images from a remote clinic in a HIPAAcompliant manner. An ophthalmologist at the clinic then used FAST to annotate 190 fundus images containing exudates at the pixelwise level in a time-efficient manner. In comparison with publicly available datasets, our dataset constitutes the largest pixelwise-labeled collection of images and the first exudate segmentation dataset with eye-matched pairs of images for a given patient. Second, we developed an optic disk CAD segmentation algorithm, where our algorithm achieved a mean intersection over union of 0.930, comparable to the disagreement between ophthalmologist annotations. We deployed this algorithm into FAST, where it segments and flushes the segmentation onto the computer screen while simultaneously filling out specified optic disk fields of a DICOM-SR report on the fundus image. Third, we integrated our software with the open-source EHR framework OpenMRS, where our software can upload both automatic and manual analyses of the fundus to a remote server using HL7 FHIR standard then retrieve historical reports for a patient chronologically. Finally, we discuss our design decisions in developing FAST, particularly those relating to its treatment of DICOM-SR reports based on fundus images and its usage of the FHIR standard and its next steps towards enabling effective analyses of fundus images.

Exudates are one of the abnormalities present in the eye which can lead to vision loss. Fundus images may consist of artifacts which occur during image acquisition and hamper the accuracy of detection of exudates. There is a need to develop an image processing based techniques for automated and correct segmentation of exudates from fundus images. This paper demonstrates an automatic computer vision algorithm for efficient identification of the exudates from fundus images by strategic fusion of techniques i.e. contrast normalization, top-hat transformation and average filtering. The proposed technique correctly detects exudates from the fundus images and rejects the artifacts and reflections. The average computation time for exudates segmentation from fundus images is 11 seconds. The proposed method is computationally efficient and robust and can be used for real time applications.