Eye Fundus Images Research Articles

InceptionResNetV2 Based Eye Cataract Detection

An eye lens is a key source for visualizing things. Sometimes it may be cloudy, caused by cataracts, and disturb the visualization process. It is necessary to detect and treat the cataract in its early stages; otherwise, it may lead to blindness. For the early detection of eye cataracts, many solutions have been proposed in past years by experts. However, it requires an expert ophthalmologist or eye surgeon to detect cataracts in its early stages because early detection of cataracts is a challenging task. Now this is the era of digitalization and AI. Everyone is talking about AI. Many researchers use AI to propose a solution to this problem. This study also proposed an AI-based cataract detection solution. A binary class eye fundus images data set was used for this research work. The ”Cataract” data set is a famous dataset for detecting cataracts on KAGGLE. Some image augmentation techniques were used to enhance the data set proficiency. In this study, InceptionResNetV2 was applied with some hyper-parameter tuning. The proposed methodology achieved a training accuracy of 99% and a validation accuracy of 98%.

Diabetic retinopathy data augmentation and vessel segmentation through deep learning based three fully convolution neural networks

ProblemThe eye fundus imaging is used for early diagnosis of most damaging concerns such as diabetic retinopathy, retinal detachments and vascular occlusions. However, the presence of noise, low contrast between background and vasculature during imaging, and vessel morphology lead to uncertain vessel segmentation. AimThis paper proposes a novel retinalblood vessel segmentation method for fundus imaging using a Difference of Gaussian (DoG) filter and an ensemble of three fully convolutional neural network (FCNN) models. MethodsA Gaussian filter with standard deviation σ1 is applied on the preprocessed grayscale fundus image and is subtracted from a similarly applied Gaussian filter with standard deviation σ2 on the same image. The resultant image is then fed into each of the three fully convolutional neural networks as the input. The FCNN models' output is then passed through a voting classifier, and a final segmented vessel structure is obtained.The Difference of Gaussian filter played an essential part in removing the high frequency details (noise) and thus finely extracted the blood vessels from the retinal fundus with underlying artifacts. ResultsThe total dataset consists of 3832 augmented images transformed from 479 fundus images. The result shows that the proposed method has performed extremely well by achieving an accuracy of 96.50%, 97.69%, and 95.78% on DRIVE, CHASE,and real-time clinical datasets respectively. ConclusionThe FCNN ensemble model has demonstrated efficacy in precisely detecting retinal vessels and in the presence of various pathologies and vasculatures.

Evaluating the performance of a non-uniform squash function in Capsule networks for early diabetic retinopathy detection using fundus image analysis

Diabetic retinopathy, known as one of the most crucial neurovascular barriers to diabetes, has affected the retina, leading to a potential risk of blindness. The harm is not isolated to personal health, with negative effects felt at the economic and social level in entire communities. The consequences of vision loss secondary to diabetic retinopathy on quality of life, independent living, and employment can be significant. It is necessary to adopt a comprehensive approach for effective management of underlying diabetes, which includes early detection, timely intervention, supportive care, and management of comorbid conditions. Critical to this is a precise classification of its stages for disease analysis. A manual diagnosis, on the other hand, requires qualified individuals to recognize the subtle nuances of the condition, making it a lengthy process. This study aims to develop a computer-assisted system for analyzing retinal fundus images to identify and classify different stages of diabetic retinopathy. A dataset of eye fundus images was used by the proposed network to classify the presence and severity of blindness and its stages. The proposed model uses a non-uniform squash function in capsule neural networks. The motive behind employing a non-uniform squash function was to encourage the model to allocate greater attention and stability to intricate images throughout the training phase. The proposed model's efficiency is assessed meticulously with pre-trained CNN methods like InceptionV3, VGG16, VGG19, and Xception. The proposed capsule network model has been revamped to address the shortcomings of prior research with care. The method stands out with an impressive 99.84 % training accuracy and 98.68 % validation accuracy. Diabetic retinopathy, though rare in occurrence, poses significant challenges due to the high cost of diagnosis, rendering it inaccessible to rural and remote communities with limited access to tests and check-ups.

An Assessment of Contemporary Methods and Data-Enabled Approaches for Early Cataract Detection

Background: Cataracts are common causes of visual impairment. Preventing blindness requires an early and accurate diagnosis. This review examines current cataract diagnosis strategies, explores data-driven machine learning algorithms for early detection, investigates the use of artificial intelligence (AI) approaches, assesses improvements in cataract detection accuracy, identifies research gaps, and provides recommendations for future studies. Methods: We gathered labelled cataract and non-cataract fundus data from the Kaggle. Scholarly publications were sourced from reliable databases such as ProQuest, IEEE, ELSEVIER, Google Scholar, and PubMed. A detailed literature search with specific terms expanded the scope of this review. We included studies that used cataract and non-cataract fundus eye images from cross-sectional, retrospective, and prospective studies. The quality assessment used the AMSTAR tool, considering factors such as literature search comprehensiveness, study selection criteria, data extraction methodologies, and study validity (Table 1). Results: This study encompassed 130 research publications, focusing on machine learning models and clinical-based diagnostic approaches for early-stage cataract identification. The performance of machine-learning models is influenced by factors such as dataset noise and limited reliable data. Barriers to the successful implementation of AI for cataract diagnosis were identified. Conclusions: This review emphasises the obstacles hindering the broad application of AI in cataract diagnosis. Addressing these findings is vital for developing strategies to overcome these challenges and enhance cataract detection systems. To achieve improved accuracy and efficiency in cataract diagnosis, future research should prioritise efforts to enhance dataset availability and quality, reduce data noise, and refine machine-learning algorithms. Unlocking the full potential of AI and/or machine learning can lead to significant breakthroughs in cataract diagnosis, ultimately resulting in better patient outcomes and reduced visual impairments.

Two-stage deep learning classification for diabetic retinopathy using gradient weighted class activation mapping

The fundus images of patients with Diabetic Retinopathy (DR) often display numerous lesions scattered across the retina. Current methods typically utilize the entire image for network learning, which has limitations since DR abnormalities are usually localized. Training Convolutional Neural Networks (CNNs) on global images can be challenging due to excessive noise. Therefore, it's crucial to enhance the visibility of important regions and focus the recognition system on them to improve accuracy. This study investigates the task of classifying the severity of diabetic retinopathy in eye fundus images by employing appropriate preprocessing techniques to enhance image quality. We propose a novel two-branch attention-guided convolutional neural network (AG-CNN) with initial image preprocessing to address these issues. The AG-CNN initially establishes overall attention to the entire image with the global branch and then incorporates a local branch to compensate for any lost discriminative cues. We conduct extensive experiments using the APTOS 2019 DR dataset. Our baseline model, DenseNet-121, achieves average accuracy/AUC values of 0.9746/0.995, respectively. Upon integrating the local branch, the AG-CNN improves the average accuracy/AUC to 0.9848/0.998, representing a significant advancement in state-of-the-art performance within the field.

Multivariate data binning and examples generation to build a Diabetic Retinopathy classifier based on temporal clinical and analytical risk factors

In this paper, we explore the possibility of exploiting retrospective clinical data from Electronic Health Records (EHR) for classification tasks in chronic patients. The different intervals, short length and high class imbalance make it unfeasible to use traditional time series techniques. The first contribution of the paper is a preprocessing method to construct a multivariate time series dataset using EHR data, which infers missing data and regularizes the data frequency. The second contribution addresses class imbalance by using domain knowledge and existing short EHR series. We synthetically extrapolate patients’ data by using similar long time series and a fuzzy-based approach. The paper addresses the problem of detection of Diabetic Retinopathy (DR). Expert domain knowledge from ophthalmologists has been used in the proposed techniques to guide the processing of time series. The novelty in that case study consists in not using eye-fundus image analysis. Instead, the proposed methods are based solely on EHR data. Several multivariate multiclass time series classifiers are used to detect the four levels of DR severity from the pre-processed data sequences. Experiments prove the quality of the sequence preprocessing techniques proposed for EHR data. Results indicate that the TapNet classifier is the best one for DR grading. Despite being tested for DR detection, the proposed data preparation methods are applicable to other diseases with similar characteristics.

Blood vitamin A level prediction in Japanese black cattle based on chromatic and dynamic eye features using double imaging system

Proactive dietary control of blood vitamin A levels is crucial for the intramuscular fat development in cattle worldwide. However, cattle become susceptible to either vitamin A deficiency or excessive state during fattening stage, influencing cattle performance, health, and beef quality. A good understanding and modelling of vitamin A levels throughout the whole cattle growth phase is needed. This study aims to assist in controlling the fattening process for production of high-marbling beef through a non-invasive monitoring of blood vitamin A levels. Using an automatic double imaging system, this study captured both surface and fundus images of cattle eyes, and based on this, predicted blood vitamin A levels through a novel dynamic analysis of 29 eye features. The best PLS model had a prediction of R2 = 0.82 and RMSE = 6.50 IU·dL−1 (equivalent to 0.02 μg · mL−1), which is of a clinically meaningful accuracy. This system can greatly facilitate vitamin A levels management in cattle raising, contributing to the effective control of beef marbling for both the market and industry.

Combining EfficientNet with ML-Decoder classification head for multi-label retinal disease classification

Retinal diseases that are not treated in time can cause irreversible, permanent damage, including blindness. Although a patient may suffer from more than one retinal disease at the same time, most of the studies focus on the diagnosis of a single disease only. Therefore, to detect multi-label retinal diseases from color fundus images, we developed an end-to-end deep learning architecture that combines the EfficientNet backbone with the ML-Decoder classification head in this study. While EfficientNet provides powerful feature extraction with fewer parameters via compound scaling, ML-Decoder further improves efficiency and flexibility by reducing quadratic dependency to a linear one and using a group decoding scheme. Also, with the use of sharpness-aware minimization (SAM) optimizer, which minimizes loss value and loss sharpness simultaneously, higher accuracy rates have been reached. In addition, a significant increase in EfficientNet performance is achieved by using image transformations and concatenation together. During the training phase, the random application of the image transformations allows for increasing the image diversity and makes the model more robust. Besides, fusing fundus images of left and right eyes at the pixel level extracts useful information about their relationship. The performance of the final model was evaluated on the publicly available Ocular Disease Intelligent Recognition (ODIR) dataset consisting of 10,000 fundus images, and superior results were obtained in all test set scenarios and performance metrics than state-of-the-art methods. The best results we obtained in the threefold cross-validation scenario for the kappa, F1, and AUC scores are 68.96%, 92.48%, and 94.80%, respectively. Moreover, it can be considered attractive in terms of floating point operations per second (FLOP) and a number of parameters.

Classification of Diabetic Retinopathy using Deep Learning

Abstract: Diabetic retinopathy, a potentially fatal retinal disease, occurs in individuals with diabetes, causing vision damage and possible visual impairment traditionally, screening for diabetic retinopathy has been the norm hard by ophthalmologists manually. To facilitate this, we have DR. Our pre-trained model is trained on a large dataset of about 3662 training images, enabling it to automatically detect the DR platform in high-resolution fundus images The dataset used for this is publicly available on kaggle. The DR phases are categorized as 0, 1, 2, 3, and 4. The fundus eye images serve as the input parameters of the model. The pre-trained model then identifies the point of interest in bank eye images, then provides activation function insights By calculating weights, for example patients’ visual intensity levels, it helps classify diabetic retinopathy images into their intensity groups appropriately within the possibility of test- methods Provides variation.

Efficacy and safety of artificial tears containing Artemia salina extract with dinucleotides for dry eye

ABSTRACT Clinical relevance This clinical trial was conducted as part of the marketing procedures for a medical device comprising artificial tears containing Artemia salina extract with dinucleotides. These molecules previously demonstrated secretagogue properties by enhancing the production of aqueous, mucinous, and lipidic components of the tears. Background After confirming the efficacy of artificial tears containing Artemia salina extract in an animal model, this study proceeded to evaluate their efficacy and safety on dry eye participants. Methods A randomised controlled clinical trial was performed on 36 dry eye participants (41.6 ± 20.6 years). Half of the participants were treated with saline solution as a placebo for four weeks, while the other half were treated with artificial tears containing Artemia salina, randomly assigned. After a wash-out period of two weeks, the treatments were crossed for another four weeks. Participants were assessed at baseline and after one week, two weeks, and four weeks. Efficacy variables were: eye dryness frequency (primary), eye comfort, visual satisfaction, tear secretion, tear break-up time, corneal staining, conjunctival staining, and conjunctival hyperaemia. Safety variables were: high- and low-contrast visual acuity, intraocular pressure, and eye fundus images analysis. Results Compared with the baseline, the saline solution showed no significant changes in any of the studied variables after four weeks of treatment (p ≥ 0.05). However, the topical instillation of the artificial tears with Artemia salina for four weeks significantly improved eye dryness frequency (p = 0.014) and corneal staining (p = 0.010). No systemic or ocular adverse events were reported during the clinical trial. Conclusion The topical instillation of artificial tears containing Artemia salina in mild to moderate dry eye participants for four weeks slightly improved their symptoms related to eye dryness frequency and reduced corneal damage, with no undesirable side effects observed.

Robust deep learning for eye fundus images: Bridging real and synthetic data for enhancing generalization

Deep learning applications for assessing medical images are limited because the datasets are often small and imbalanced. The use of synthetic data has been proposed in the literature, but neither a robust comparison of the different methods nor generalizability has been reported. Our approach integrates a retinal image quality assessment model and StyleGAN2 architecture to enhance Age-related Macular Degeneration (AMD) detection capabilities and improve generalizability. This work compares ten different Generative Adversarial Network (GAN) architectures to generate synthetic eye-fundus images with and without AMD. We combined subsets of three public databases (iChallenge-AMD, ODIR-2019, and RIADD) to form a single training and test set. We employed the STARE dataset for external validation, ensuring a comprehensive assessment of the proposed approach. The results show that StyleGAN2 reached the lowest Fréchet Inception Distance (166.17), and clinicians could not accurately differentiate between real and synthetic images. ResNet-18 architecture obtained the best performance with 85% accuracy and outperformed the two human experts (80% and 75%) in detecting AMD fundus images. The accuracy rates were 82.8% for the test set and 81.3% for the STARE dataset, demonstrating the model’s generalizability. The proposed methodology for synthetic medical image generation has been validated for robustness and accuracy, with free access to its code for further research and development in this field.

Topology Regulated Background Extraction (TRBE) method for eye fundus images

One of the initial steps in the preprocessing of digital fundoscopy images is the identification of pixels containing relevant information. This can be achieved through different approaches, one of them is implementing background extraction, reducing the set of pixels to be analyzed later in the process. In this work, we present a background extraction method for digital fundoscopy images based on computational topology. By interpreting binarized images as cubical complexes and extracting their homological groups in 1 and 2 dimensions we identify a subset of luminescence values that can be used to binarize the original grayscale image, obtaining a mask to achieve background extraction. This method is robust to noise and suboptimal image quality, facilitating the analytical pipeline in the context of computer aided diagnosis approaches. This method facilitates the segmentation of the background of a digital fundoscopy image, which allows further methods to focus on pixels with relevant information (eye fundus). This tool is best suited to be implemented in the preprocessing stages of the analytical pipeline by computational ophthalmology specialists.•It is robust to noise and low-quality images.•Output provides an ideal scenario for down-the-line analysis by facilitating only relevant pixels in a digital fundoscopy.

Progress and challenges of in vivo flow cytometry and its applications in circulating cells of eyes.

Circulating inflammatory cells in eyes have emerged as early indicators of numerous major diseases, yet the monitoring of these cells remains an underdeveloped field. In vivo flow cytometry (IVFC), a noninvasive technique, offers the promise of real-time, dynamic quantification of circulating cells. However, IVFC has not seen extensive applications in the detection of circulating cells in eyes, possibly due to the eye's unique physiological structure and fundus imaging limitations. This study reviews the current research progress in retinal flow cytometry and other fundus examination techniques, such as adaptive optics, ultra-widefield retinal imaging, multispectral imaging, and optical coherence tomography, to propose novel ideas for circulating cell monitoring.

Eye Disease Classification Based On Fundus Image Using Yolo V8 Algorithm

Eye disease is a very serious problem because it affects one of the five human senses. In many cases many people ignore the impact of eye diseases in the early stages. In general, the process of examining eye diseases is carried out based on manual analysis by doctors (experts) on the fundus image of the patient's eye at a fairly expensive cost. To overcome this, the author proposes an eye disease classification system that can automatically detect eye diseases using YOLO V8. This system can be used for early detection of eye diseases to prevent the development of more serious eye diseases. From the test results of the model built, the accuracy value is 92%, precision is 91%, recall is 92%, F1-score is 91%. Overall, these results can be considered satisfactory and can be implemented for eye disease classification systems based on fundus images.

Improved predictive diagnosis of diabetic macular edema based on hybrid models: An observational study

Diabetic Macular Edema (DME) is the most common sight-threatening complication of type 2 diabetes. Optical Coherence Tomography (OCT) is the most useful imaging technique to diagnose, follow up, and evaluate treatments for DME. However, OCT exam and devices are expensive and unavailable in all clinics in low- and middle-income countries. Our primary goal was therefore to develop an alternative method to OCT for DME diagnosis by introducing spectral information derived from spontaneous electroretinogram (ERG) signals as a single input or combined with fundus that is much more widespread. Baseline ERGs were recorded in 233 patients and transformed into scalograms and spectrograms via Wavelet and Fourier transforms, respectively. Using transfer learning, distinct Convolutional Neural Networks (CNN) were trained as classifiers for DME using OCT, scalogram, spectrogram, and eye fundus images. Input data were randomly split into training and test sets with a proportion of 80 %–20 %, respectively. The top performers for each input type were selected, OpticNet-71 for OCT, DenseNet-201 for eye fundus, and non-evoked ERG-derived scalograms, to generate a combined model by assigning different weights for each of the selected models. Model validation was performed using a dataset alien to the training phase of the models. None of the models powered by mock ERG-derived input performed well. In contrast, hybrid models showed better results, in particular, the model powered by eye fundus combined with mock ERG-derived information with a 91 % AUC and 86 % F1-score, and the model powered by OCT and mock ERG-derived scalogram images with a 93 % AUC and 89 % F1-score. These data show that the spontaneous ERG-derived input adds predictive value to the fundus- and OCT-based models to diagnose DME, except for the sensitivity of the OCT model which remains the same. The inclusion of mock ERG signals, which have recently been shown to take only 5 min to record in daylight conditions, therefore represents a potential improvement over existing OCT-based models, as well as a reliable and cost-effective alternative when combined with the fundus, especially in underserved areas, to predict DME.

Predicting diabetic retinopathy stage using Siamese Convolutional Neural Network

ABSTRACT Diabetic retinopathy (DR) is considered to be a leading cause of blindness in people aged 16 to 64 years, affecting around 40% of the population diagnosed with diabetes mellitus (DM). DR is usually identified through retinal image analysis, and in some countries, such as Brazil, such diagnosis is hindered by limited access to specialized care, leading to lengthy waits for ophthalmological evaluations. This scenario makes the Brazilian Diabetic Society’s annual DR check recommendation impractical for many. To address this gap, our study introduces a novel Siamese Convolutional Neural Network (SCNN) for DR prediction, usable by primary care professionals. Our SCNN analyzes pairs of eye fundus images and employs shared weights in its layers to extract essential features, facilitating similarity measurement between neural network outputs. Despite challenges with limited and imbalanced datasets, our SCNN showed effectiveness. We tested it on four datasets (IDRiD, APTOS, Messidor-1, DIARETDB0), with accuracy ranging from 67.23% (APTOS) to 96.85% (DIARETDB0). Compared to other methods, our approach consistently excelled, especially in recall analysis. These results suggest that deep learning via Siamese networks is likely to be a viable and potential DR screening tool.

Glaucoma Identification Using Relevance Vector Machine

Glaucoma is a chronic optic neuropathy characterized by curvature of the optic disc and narrowing of the visual field, accompanied by increased intraocular pressure. In glaucoma there will be a weakening of eye function with visual field defects and anatomical damage in the form of excavation (reverberation) and degeneration of the optic nerve pupil which can end in blindness. The glaucoma identification process is done by eye fundus image. The process of identifying glaucoma is still done by the ophthalmologist manually by looking at the fundus image. There are already people who have done this research using the K-NN method, but the level of accuracy is still small and has not measured the processing time for glaucoma identification. Based on this, this study will apply the Relevance Vector Machine method to the glaucoma identification system on the human retina. System performance is measured based on the values of accuracy, precision, recall, and F-Measure. Experiments were carried out on the glaucoma dataset. The average values in terms of accuracy, precision, recall, and f-measure were 80% accuracy, 86% precision, 80% recall, and 79% f-measure, respectively. These values are influenced by the number of datasets from training images, validation images, and test images. Based on these results, the proposed Relevance Vector Machine has an 80% accuracy performance.

In-Depth Evaluation of Saliency Maps for Interpreting Convolutional Neural Network Decisions in the Diagnosis of Glaucoma Based on Fundus Imaging.

Glaucoma, a leading cause of blindness, damages the optic nerve, making early diagnosis challenging due to no initial symptoms. Fundus eye images taken with a non-mydriatic retinograph help diagnose glaucoma by revealing structural changes, including the optic disc and cup. This research aims to thoroughly analyze saliency maps in interpreting convolutional neural network decisions for diagnosing glaucoma from fundus images. These maps highlight the most influential image regions guiding the network's decisions. Various network architectures were trained and tested on 739 optic nerve head images, with nine saliency methods used. Some other popular datasets were also used for further validation. The results reveal disparities among saliency maps, with some consensus between the folds corresponding to the same architecture. Concerning the significance of optic disc sectors, there is generally a lack of agreement with standard medical criteria. The background, nasal, and temporal sectors emerge as particularly influential for neural network decisions, showing a likelihood of being the most relevant ranging from 14.55% to 28.16% on average across all evaluated datasets. We can conclude that saliency maps are usually difficult to interpret and even the areas indicated as the most relevant can be very unintuitive. Therefore, its usefulness as an explanatory tool may be compromised, at least in problems such as the one addressed in this study, where the features defining the model prediction are generally not consistently reflected in relevant regions of the saliency maps, and they even cannot always be related to those used as medical standards.

A Robust Machine Learning Model for Diabetic Retinopathy Classification.

Ensemble learning is a process that belongs to the artificial intelligence (AI) field. It helps to choose a robust machine learning (ML) model, usually used for data classification. AI has a large connection with image processing and feature classification, and it can also be successfully applied to analyzing fundus eye images. Diabetic retinopathy (DR) is a disease that can cause vision loss and blindness, which, from an imaging point of view, can be shown when screening the eyes. Image processing tools can analyze and extract the features from fundus eye images, and these corroborate with ML classifiers that can perform their classification among different disease classes. The outcomes integrated into automated diagnostic systems can be a real success for physicians and patients. In this study, in the form image processing area, the manipulation of the contrast with the gamma correction parameter was applied because DR affects the blood vessels, and the structure of the eyes becomes disorderly. Therefore, the analysis of the texture with two types of entropies was necessary. Shannon and fuzzy entropies and contrast manipulation led to ten original features used in the classification process. The machine learning library PyCaret performs complex tasks, and the empirical process shows that of the fifteen classifiers, the gradient boosting classifier (GBC) provides the best results. Indeed, the proposed model can classify the DR degrees as normal or severe, achieving an accuracy of 0.929, an F1 score of 0.902, and an area under the curve (AUC) of 0.941. The validation of the selected model with a bootstrap statistical technique was performed. The novelty of the study consists of the extraction of features from preprocessed fundus eye images, their classification, and the manipulation of the contrast in a controlled way.

TDCAU-Net: retinal vessel segmentation using transformer dilated convolutional attention-based U-Net method

Retinal vessel segmentation plays a vital role in the medical field, facilitating the identification of numerous chronic conditions based on retinal vessel images. These conditions include diabetic retinopathy, hypertensive retinopathy, glaucoma, and others. Although the U-Net model has shown promising results in retinal vessel segmentation, it tends to struggle with fine branching and dense vessel segmentation. To further enhance the precision of retinal vessel segmentation, we propose a novel approach called transformer dilated convolution attention U-Net (TDCAU-Net), which builds upon the U-Net architecture with improved Transformer-based dilated convolution attention mechanisms. The proposed model retains the three-layer architecture of the U-Net network. The Transformer component enables the learning of contextual information for each pixel in the image, while the dilated convolution attention prevents information loss. The algorithm efficiently addresses several challenges to optimize blood vessel detection. The process starts with five-step preprocessing of the images, followed by chunking them into segments. Subsequently, the retinal images are fed into the modified U-Net network introduced in this paper for segmentation. The study employs eye fundus images from the DRIVE and CHASEDB1 databases for both training and testing purposes. Evaluation metrics are utilized to compare the algorithm’s results with state-of-the-art methods. The experimental analysis on both databases demonstrates that the algorithm achieves high values of sensitivity, specificity, accuracy, and AUC. Specifically, for the first database, the achieved values are 0.8187, 0.9756, 0.9556, and 0.9795, respectively. For the second database, the corresponding values are 0.8243, 0.9836, 0.9738, and 0.9878, respectively. These results demonstrate that the proposed approach outperforms state-of-the-art methods, achieving higher performance on both datasets. The TDCAU-Net model presented in this study exhibits substantial capabilities in accurately segmenting fine branching and dense vessels. The segmentation performance of the network surpasses that of the U-Net algorithm and several mainstream methods.