Color Fundus Images Research Articles

A Regression-Based Approach to Diabetic Retinopathy Diagnosis Using Efficientnet.

The aim of this study is to develop a computer-assisted solution for the efficient and effective detection of diabetic retinopathy (DR), a complication of diabetes that can damage the retina and cause vision loss if not treated in a timely manner. Manually diagnosing DR through color fundus images requires a skilled clinician to spot lesions, but this can be challenging, especially in areas with a shortage of trained experts. As a result, there is a push to create computer-aided diagnosis systems for DR to help reduce the time it takes to diagnose the condition. The detection of diabetic retinopathy through automation is challenging, but convolutional neural networks (CNNs) play a vital role in achieving success. CNNs have been proven to be more effective in image classification than methods based on handcrafted features. This study proposes a CNN-based approach for the automated detection of DR using Efficientnet-B0 as the backbone network. The authors of this study take a unique approach by viewing the detection of diabetic retinopathy as a regression problem rather than a traditional multi-class classification problem. This is because the severity of DR is often rated on a continuous scale, such as the international clinical diabetic retinopathy (ICDR) scale. This continuous representation provides a more nuanced understanding of the condition, making regression a more suitable approach for DR detection compared to multi-class classification. This approach has several benefits. Firstly, it allows for more fine-grained predictions as the model can assign a value that falls between the traditional discrete labels. Secondly, it allows for better generalization. The model was tested on the APTOS and DDR datasets. The proposed model demonstrated improved efficiency and accuracy in detecting DR compared to traditional methods. This method has the potential to enhance the efficiency and accuracy of DR diagnosis, making it a valuable tool for healthcare professionals. The model has the potential to aid in the rapid and accurate diagnosis of DR, leading to the improved early detection, and management, of the disease.

Automatic vessel crossing and bifurcation detection based on multi-attention network vessel segmentation and directed graph search

Analysis of the vascular tree is the basic premise to automatically diagnose retinal biomarkers associated with ophthalmic and systemic diseases, among which accurate identification of intersection and bifurcation points is quite challenging but important for disentangling complex vascular network and tracking vessel morphology. In this paper, we present a novel directed graph search-based multi-attentive neural network approach to automatically segment the vascular network and separate intersections and bifurcations from color fundus images. Our approach uses multi-dimensional attention to adaptively integrate local features and their global dependencies while learning to focus on target structures at different scales to generate binary vascular maps. A directed graphical representation of the vascular network is constructed to represent the topology and spatial connectivity of the vascular structures. Using local geometric information including color difference, diameter, and angle, the complex vascular tree is decomposed into multiple sub-trees to finally classify and label vascular feature points. The proposed method has been tested on the DRIVE dataset and the IOSTAR dataset containing 40 images and 30 images, respectively, with 0.863 and 0.764 F1-score of detection points and average accuracy of 0.914 and 0.854 for classification points. These results demonstrate the superiority of our proposed method outperforming state-of-the-art methods in feature point detection and classification.

Photostress Recovery Time as a Potential Predictive Biomarker for Age-Related Macular Degeneration.

The purpose of this study was to assess recovery time following photostress and its association with age-related macular degeneration (AMD) cross-sectionally and longitudinally in an elderly population-based cohort. We analyzed photostress recovery time (PRT) and AMD in >1800 AugUR study participants aged 70+ years. On color fundus images from baseline and 3-year follow-up, presence of AMD was graded manually (Three Continent AMD Consortium Severity Scale). Visual acuity (VA) was assessed via Early Treatment Diabetic Retinopathy Study (ETDRS) charts. After a 30-second bleaching of the macular region via direct ophthalmoscope, PRT was measured as the seconds to regain VA. First, we analyzed 1208 AugUR participants cross-sectionally (288 with early AMD, and 78 with late AMD). Prolonged PRT was associated with early and late AMD versus no AMD (median PRT = 119.5, 198.0 versus 80.0 seconds, respectively; logistic regression odds ratio [OR] = 1.109-1.165 per 10 seconds, P values < 0.0001). Sensitivity analyses using alternative models or restricting to participants after cataract surgery revealed similar ORs. Second, the association was confirmed in an independent cross-sectional AugUR sample (n = 486). Third, in longitudinal analysis of 233 AugUR participants without AMD, prolonged PRT was associated with incident AMD ascertained 3 years later (follow-up time = 3.2 ± 0.2 years, OR = 1.112-1.162 per 10 seconds, P < 0.05). Overall, we demonstrate a significant association of prolonged PRT with AMD cross-sectionally and longitudinally in elderly individuals. Prolonged PRT might capture retinal function impairment after cell damage before early AMD is visible via color fundus imaging. Our results suggest PRT as quantitative predictive biomarker for incident AMD, making it potentially worthwhile also for clinical care.

Chákṣu: A glaucoma specific fundus image database

We introduce Chákṣu–a retinal fundus image database for the evaluation of computer-assisted glaucoma prescreening techniques. The database contains 1345 color fundus images acquired using three brands of commercially available fundus cameras. Each image is provided with the outlines for the optic disc (OD) and optic cup (OC) using smooth closed contours and a decision of normal versus glaucomatous by five expert ophthalmologists. In addition, segmentation ground-truths of the OD and OC are provided by fusing the expert annotations using the mean, median, majority, and Simultaneous Truth and Performance Level Estimation (STAPLE) algorithm. The performance indices show that the ground-truth agreement with the experts is the best with STAPLE algorithm, followed by majority, median, and mean. The vertical, horizontal, and area cup-to-disc ratios are provided based on the expert annotations. Image-wise glaucoma decisions are also provided based on majority voting among the experts. Chákṣu is the largest Indian-ethnicity-specific fundus image database with expert annotations and would aid in the development of artificial intelligence based glaucoma diagnostics.

Dual Consistency Enabled Weakly and Semi-Supervised Optic Disc and Cup Segmentation with Dual Adaptive Graph Convolutional Networks.

Glaucoma is a progressive eye disease that results in permanent vision loss, and the vertical cup to disc ratio (vCDR) in colour fundus images is essential in glaucoma screening and assessment. Previous fully supervised convolution neural networks segment the optic disc (OD) and optic cup (OC) from color fundus images and then calculate the vCDR offline. However, they rely on a large set of labeled masks for training, which is expensive and time-consuming to acquire. To address this, we propose a weakly and semi-supervised graph-based network that investigates geometric associations and domain knowledge between segmentation probability maps (PM), modified signed distance function representations (mSDF), and boundary region of interest characteristics (B-ROI) in three aspects. Firstly, we propose a novel Dual Adaptive Graph Convolutional Network (DAGCN) to reason the long-range features of the PM and the mSDF w.r.t. the regional uniformity. Secondly, we propose a dual consistency regularization-based semi-supervised learning paradigm. The regional consistency between the PM and the mSDF, and the marginal consistency between the derived B-ROI from each of them boost the proposed model's performance due to the inherent geometric associations. Thirdly, we exploit the task-specific domain knowledge via the oval shapes of OD & OC, where a differentiable vCDR estimating layer is proposed. Furthermore, without additional annotations, the supervision on vCDR serves as weakly-supervisions for segmentation tasks. Experiments on six large-scale datasets demonstrate our model's superior performance on OD & OC segmentation and vCDR estimation. The implementation code has been made available 1.

Automatic Multilabel Classification of Multiple Fundus Diseases Based on Convolutional Neural Network With Squeeze-and-Excitation Attention.

Automatic multilabel classification of multiple fundus diseases is of importance for ophthalmologists. This study aims to design an effective multilabel classification model that can automatically classify multiple fundus diseases based on color fundus images. We proposed a multilabel fundus disease classification model based on a convolutional neural network to classify normal and seven categories of common fundus diseases. Specifically, an attention mechanism was introduced into the network to further extract information features from color fundus images. The fundus images with eight categories of labels were applied to train, validate, and test our model. We employed the validation accuracy, area under the receiver operating characteristic curve (AUC), and F1-score as performance metrics to evaluate our model. Our proposed model achieved better performance with a validation accuracy of 94.27%, an AUC of 85.80%, and an F1-score of 86.08%, compared to two state-of-the-art models. Most important, the number of training parameters has dramatically dropped by three and eight times compared to the two state-of-the-art models. This model can automatically classify multiple fundus diseases with not only excellent accuracy, AUC, and F1-score but also significantly fewer training parameters and lower computational cost, providing a reliable assistant in clinical screening. The proposed model can be widely applied in large-scale multiple fundus disease screening, helping to create more efficient diagnostics in primary care settings.

Combining convolutional neural networks and self-attention for fundus diseases identification

Early detection of lesions is of great significance for treating fundus diseases. Fundus photography is an effective and convenient screening technique by which common fundus diseases can be detected. In this study, we use color fundus images to distinguish among multiple fundus diseases. Existing research on fundus disease classification has achieved some success through deep learning techniques, but there is still much room for improvement in model evaluation metrics using only deep convolutional neural network (CNN) architectures with limited global modeling ability; the simultaneous diagnosis of multiple fundus diseases still faces great challenges. Therefore, given that the self-attention (SA) model with a global receptive field may have robust global-level feature modeling ability, we propose a multistage fundus image classification model MBSaNet which combines CNN and SA mechanism. The convolution block extracts the local information of the fundus image, and the SA module further captures the complex relationships between different spatial positions, thereby directly detecting one or more fundus diseases in retinal fundus image. In the initial stage of feature extraction, we propose a multiscale feature fusion stem, which uses convolutional kernels of different scales to extract low-level features of the input image and fuse them to improve recognition accuracy. The training and testing were performed based on the ODIR-5k dataset. The experimental results show that MBSaNet achieves state-of-the-art performance with fewer parameters. The wide range of diseases and different fundus image collection conditions confirmed the applicability of MBSaNet.

Spatial Correlation Module for Classification of Multi-Label Ocular Diseases Using Color Fundus Images

To prevent irreversible damage to one’s eyesight, ocular diseases (ODs) need to be recognized and treated immediately. Color fundus imaging (CFI) is a screening technology that is both effective and economical. According to CFIs, the early stages of the disease are characterized by a paucity of observable symptoms, which necessitates the prompt creation of automated and robust diagnostic algorithms. The traditional research focuses on image-level diagnostics that attend to the left and right eyes in isolation without making use of pertinent correlation data between the two sets of eyes. In addition, they usually only target one or a few different kinds of eye diseases at the same time. In this study, we design a patient-level multi-label OD (PLML_ODs) classification model that is based on a spatial correlation network (SCNet). This model takes into consideration the relevance of patient-level diagnosis combining bilateral eyes and multi-label ODs classification. PLML_ODs is made up of three parts: a backbone convolutional neural network (CNN) for feature extraction i.e., DenseNet-169, a SCNet for feature correlation, and a classifier for the development of classification scores. The DenseNet-169 is responsible for retrieving two separate sets of attributes, one from each of the left and right CFI. After then, the SCNet will record the correlations between the two feature sets on a pixel-by-pixel basis. After the attributes have been analyzed, they are integrated to provide a representation at the patient level. Throughout the whole process of ODs categorization, the patient-level representation will be used. The efficacy of the PLML_ODs is examined using a soft margin loss on a dataset that is readily accessible to the public, and the results reveal that the classification performance is significantly improved when compared to several baseline approaches.

Diabetic Retinopathy Grading From Color Fundus Images: An Autotuned Deep Learning Approach

Diabetic Retinopathy (DR) is a consequence of long-term diabetes which affects the eyes. It causes affliction to the veins in the eyes causing raptures on the retina which can impact vision. If the condition is failed to be detected at an early stage, it can lead to complete vision loss. The conventional diagnosing cycle of DR using fundus images requires very skilled practitioners due to the minute nature of features of the anomalies which can even lead to misdiagnosis and is time consuming. Therefore, devising an automated method for the diagnosis of DR can assist individuals with diabetes to identify symptoms of DR at a very early stage. This research also encompasses the classification of the detected images into its corresponding 3 stages namely No Diabetic Retinopathy (No DR), Non-Proliferative Diabetic Retinopathy (NPDR) and Proliferative Diabetic Retinopathy (PDR) that can greatly aid in monitoring the dynamics of the key features including lesions, hemorrhages, and density of blood vessels. Deep learning algorithms has gotten to become a well-known method that can accomplish a wide variety of classification tasks. However, most of these methods are only efficient in classifying the various stages of DR with low accuracies notably, for the early stages. The devised algorithm in this research uses enriched images processing techniques, automatic hyperparameter tuning and neural network training strategies to provide more emphasis on the minute features for better prediction. The algorithm was tested and compared with modified Resnet50, VGG16, Mobilenetv2, Inceptionv3 and InceptionResnetv2 which gave a classification accuracy of 94.7%, 86.1%, 85.8%, 85.3% and 87% respectively with corresponding detection accuracies of 99.8%, 94%, 94.2%, 94.9% and 98.2% respectively on a test set of 508 images. Using the proposed algorithm, the results indicate Resnet50 based network gave superior performance for both detection and classification tasks.

Hybrid multi-kernel SVM algorithm for microaneurysm recognition in colour fundus images

Hybrid multi-kernel SVM algorithm for microaneurysm recognition in colour fundus images

Use of Segment Anything Model (SAM) and MedSAM in the optic disc Segmentation of colour retinal fundus images: Experimental Finding

Detection of Optic Disc segmentation in retinal fundus images is important step in identification of various abnormal conditions like diabetic retinopathy, Glaucoma etc. and is an important part of eye that is routinely examined. We have used Segment Anything model (SAM) by Meta AI and based fine-tuned model MedSAM for segmentation of optic disc in retinal fundus images. We have used Indian Diabetic Retinopathy dataset (IDRiD) segmentation part. It consists of 81 original colour fundus images in jpg files split into train and test set. Ground-truth images for the lesions (Microaneurysms, Haemorrhages, Hard Exudates and Soft Exudates divided into train and test set - TIF Files) and Optic Disc (divided into train and test set - TIF Files). This dataset is feed into the deep learning neural network model; Segment Anything Model (SAM) and MedSAM for the image segmentation task. The Dice Similarity Coefficient (DSC) is used in the experiment for observing the model performance. We have used 80% for training and 20% for testing of Indian Diabetic Retinopathy Dataset (IDRiD) in both models. After 100 epochs with 32 batch size it is observed that average Dice Similarity Coefficient (DSC) of SAM and MedSAM models in the optic disc segmentation task are, 85.97 % and 90.15 %. Also, in finding it is observed that SAM and MedSAM both have very low DSC where the fundus images are having very high brightness.

Deployment of CNN on colour fundus images for the automatic detection of glaucoma

Deployment of CNN on colour fundus images for the automatic detection of glaucoma

Use of Segment Anything Model (SAM) and MedSAM in the optic disc Segmentation of colour retinal fundus images: Experimental Finding

Detection of Optic Disc segmentation in retinal fundus images is important step in identification of various abnormal conditions like diabetic retinopathy, Glaucoma etc. and is an important part of eye that is routinely examined. We have used Segment Anything model (SAM) by Meta AI and based fine-tuned model MedSAM for segmentation of optic disc in retinal fundus images. We have used Indian Diabetic Retinopathy dataset (IDRiD) segmentation part. It consists of 81 original colour fundus images in jpg files split into train and test set. Ground-truth images for the lesions (Microaneurysms, Haemorrhages, Hard Exudates and Soft Exudates divided into train and test set - TIF Files) and Optic Disc (divided into train and test set - TIF Files). This dataset is feed into the deep learning neural network model; Segment Anything Model (SAM) and MedSAM for the image segmentation task. The Dice Similarity Coefficient (DSC) is used in the experiment for observing the model performance. We have used 80% for training and 20% for testing of Indian Diabetic Retinopathy Dataset (IDRiD) in both models. After 100 epochs with 32 batch size it is observed that average Dice Similarity Coefficient (DSC) of SAM and MedSAM models in the optic disc segmentation task are, 85.97 % and 90.15 %. Also, in finding it is observed that SAM and MedSAM both have very low DSC where the fundus images are having very high brightness.

Use of Segment Anything Model (SAM) and MedSAM in the optic disc Segmentation of colour retinal fundus images: Experimental Finding

Detection of Optic Disc segmentation in retinal fundus images is important step in identification of various abnormal conditions like diabetic retinopathy, Glaucoma etc. and is an important part of eye that is routinely examined. We have used Segment Anything model (SAM) by Meta AI and based fine-tuned model MedSAM for segmentation of optic disc in retinal fundus images. We have used Indian Diabetic Retinopathy dataset (IDRiD) segmentation part. It consists of 81 original colour fundus images in jpg files split into train and test set. Ground-truth images for the lesions (Microaneurysms, Haemorrhages, Hard Exudates and Soft Exudates divided into train and test set - TIF Files) and Optic Disc (divided into train and test set - TIF Files). This dataset is feed into the deep learning neural network model; Segment Anything Model (SAM) and MedSAM for the image segmentation task. The Dice Similarity Coefficient (DSC) is used in the experiment for observing the model performance. We have used 80% for training and 20% for testing of Indian Diabetic Retinopathy Dataset (IDRiD) in both models. After 100 epochs with 32 batch size it is observed that average Dice Similarity Coefficient (DSC) of SAM and MedSAM models in the optic disc segmentation task are, 85.97 % and 90.15 %. Also, in finding it is observed that SAM and MedSAM both have very low DSC where the fundus images are having very high brightness.

Diabetic Retinopathy Detection Through Deep Learning Using CNN Wide-Net-X architecture

In recent times, Diabetic Retinopathy (DR) has emerged as a critical complication for patients with diabetes, where the blood vessels in the retina are severely damaged, potentially leading to vision loss and, if left untreated, blindness. The World Health Organization has projected that by 2040, DR will impact around 224 million people. To address this issue, this research paper proposes CNN Wide-Net-X architecture model for image classification, which utilizes colour fundus images to detect Diabetic Retinopathy. The objective of this model is to enhance the accuracy and efficiency of the diagnostic process. For training and testing the model, the EyePACS dataset consisting of 5220 images is utilized, which is a widely accepted dataset for detecting Diabetic Retinopathy. To evaluate the performance of our model, we use metrics such as accuracy, precision, recall, and F1-score. The proposed CNN model is a significant step towards early detection and accurate diagnosis of DR. It is hoped that with the increased accuracy and efficiency provided by this model, patients with DR can receive timely treatment, thereby reducing the risk of vision loss and blindness.

Gaussian Blur Masked ResNet2.0 Architecture for Diabetic Retinopathy燚etection

Diabetic Retinopathy (DR) is a serious hazard that can result in irreversible blindness if not addressed in a timely manner. Hence, numerous techniques have been proposed for the accurate and timely detection of this disease. Out of these, Deep Learning (DL) and Computer Vision (CV) methods for multiclass categorization of color fundus images diagnosed with Diabetic Retinopathy have sparked considerable attention. In this paper, we attempt to develop an extended ResNet152V2 architecture-based Deep Learning model, named ResNet2.0 to aid the timely detection of DR. The APTOS-2019 dataset was used to train the model. This consists of 3662 fundus images belonging to five different stages of DR: no DR (Class 0), mild DR (Class 1), moderate DR (Class 2), severe DR (Class 3), and proliferative DR (Class 4). The model was gauged based on ability to detect stage-wise DR. The images were pre-processed using negative and positive weighted Gaussian-based masks as feature engineering to further enhance the quality of the fundus images by removing the noise and normalizing the images. Up-sampling and data augmentation methods were used to address the skewness of the original dataset. The proposed model achieved an overall accuracy of 91% and an area under the receiver-operating characteristic curve (AUC) score of 95.1%, outperforming existing Deep Learning models by around 10%. Furthermore, the class-wise F1 score for No DR was 92%, Mild DR was 82%, Moderate DR was 66%, Severe was DR 89% and Proliferative DR was 80%.

Lesion-aware generative adversarial networks for color fundus image to fundus fluorescein angiography translation

Background and Objective: Fundus fluorescein angiography (FFA) is widely used in clinical ophthalmic diagnosis and treatment with the requirement of adverse fluorescent dyes injection. Recently, many deep Convolutional Neural Network(CNN)-based methods have been proposed to estimate FFA from color fundus (CF) images to eliminate the use of adverse fluorescent dyes. However, the robustness of these methods is affected by pathological changes.Method: In this work, we present a CNN-based approach, lesion-aware generative adversarial networks (LA-GAN), to enhance the visual effect of lesion characteristics in the generated FFA images. First, we lead the generator notice lesion information by joint learning with lesion region segmentation. A new hierarchical correlation multi-task framework for high-resolution images is designed. Second, to enhance the visual contrast between normal regions and lesion regions, a newly designed region-level adversarial loss is used rather than the image-level adversarial loss. The code is publicly available at: https://github.com/nicetomeetu21/LA-GAN.Results: The effectiveness of LA-Net has been verified in data with branch retinal vein occlusion. The proposed model reported as measures of generation performance a mean structural similarity (SSIM) of 0.536, mean learned perceptual image patch similarity (LPIPS) 0.312, outperforming other FFA generation and general image generation methods. Further, due to the proposed multi-task learning framework, the lesion-region segmentation performance was further reported as the mean Dice increased from 0.714 to 0.797 and the mean accuracy increased from 0.873 to 0.905, outperforming general single-task image segmentation methods.Conclusions: The results show that the visual effect of lesion characteristics can be improved by employing the region-level adversarial loss and the hierarchical correlation multi-task framework respectively. Based on the results of comparison with the state-of-the-art methods, LA-GAN is not only effective for CF-to-FFA translation, but also effective for lesion-region segmentation. Thus, it may be used for various image translation and lesion segmentation tasks in future research.

Application of a deep learning system in glaucoma screening and further classification with colour fundus photographs: a case control study

BackgroundTo verify efficacy of automatic screening and classification of glaucoma with deep learning system.MethodsA cross-sectional, retrospective study in a tertiary referral hospital. Patients with healthy optic disc, high-tension, or normal-tension glaucoma were enrolled. Complicated non-glaucomatous optic neuropathy was excluded. Colour and red-free fundus images were collected for development of DLS and comparison of their efficacy. The convolutional neural network with the pre-trained EfficientNet-b0 model was selected for machine learning. Glaucoma screening (Binary) and ternary classification with or without additional demographics (age, gender, high myopia) were evaluated, followed by creating confusion matrix and heatmaps. Area under receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, and F1 score were viewed as main outcome measures.ResultsTwo hundred and twenty-two cases (421 eyes) were enrolled, with 1851 images in total (1207 normal and 644 glaucomatous disc). Train set and test set were comprised of 1539 and 312 images, respectively. If demographics were not provided, AUC, accuracy, precision, sensitivity, F1 score, and specificity of our deep learning system in eye-based glaucoma screening were 0.98, 0.91, 0.86, 0.86, 0.86, and 0.94 in test set. Same outcome measures in eye-based ternary classification without demographic data were 0.94, 0.87, 0.87, 0.87, 0.87, and 0.94 in our test set, respectively. Adding demographics has no significant impact on efficacy, but establishing a linkage between eyes and images is helpful for a better performance. Confusion matrix and heatmaps suggested that retinal lesions and quality of photographs could affect classification. Colour fundus images play a major role in glaucoma classification, compared to red-free fundus images.ConclusionsPromising results with high AUC and specificity were shown in distinguishing normal optic nerve from glaucomatous fundus images and doing further classification.

Weakly-supervised detection of AMD-related lesions in color fundus images using explainable deep learning

Background and ObjectivesAge-related macular degeneration (AMD) is a degenerative disorder affecting the macula, a key area of the retina for visual acuity. Nowadays, AMD is the most frequent cause of blindness in developed countries. Although some promising treatments have been proposed that effectively slow down its development, their effectiveness significantly diminishes in the advanced stages. This emphasizes the importance of large-scale screening programs for early detection. Nevertheless, implementing such programs for a disease like AMD is usually unfeasible, since the population at risk is large and the diagnosis is challenging. For the characterization of the disease, clinicians have to identify and localize certain retinal lesions. All this motivates the development of automatic diagnostic methods. In this sense, several works have achieved highly positive results for AMD detection using convolutional neural networks (CNNs). However, none of them incorporates explainability mechanisms linking the diagnosis to its related lesions to help clinicians to better understand the decisions of the models. This is specially relevant, since the absence of such mechanisms limits the application of automatic methods in the clinical practice. In that regard, we propose an explainable deep learning approach for the diagnosis of AMD via the joint identification of its associated retinal lesions. MethodsIn our proposal, a CNN with a custom architectural setting is trained end-to-end for the joint identification of AMD and its associated retinal lesions. With the proposed setting, the lesion identification is directly derived from independent lesion activation maps; then, the diagnosis is obtained from the identified lesions. The training is performed end-to-end using image-level labels. Thus, lesion-specific activation maps are learned in a weakly-supervised manner. The provided lesion information is of high clinical interest, as it allows clinicians to assess the developmental stage of the disease. Additionally, the proposed approach allows to explain the diagnosis obtained by the models directly from the identified lesions and their corresponding activation maps. The training data necessary for the approach can be obtained without much extra work on the part of clinicians, since the lesion information is habitually present in medical records. This is an important advantage over other methods, including fully-supervised lesion segmentation methods, which require pixel-level labels whose acquisition is arduous. ResultsThe experiments conducted in 4 different datasets demonstrate that the proposed approach is able to identify AMD and its associated lesions with satisfactory performance. Moreover, the evaluation of the lesion activation maps shows that the models trained using the proposed approach are able to identify the pathological areas within the image and, in most cases, to correctly determine to which lesion they correspond. ConclusionsThe proposed approach provides meaningful information—lesion identification and lesion activation maps—that conveniently explains and complements the diagnosis, and is of particular interest to clinicians for the diagnostic process. Moreover, the data needed to train the networks using the proposed approach is commonly easy to obtain, what represents an important advantage in fields with particularly scarce data, such as medical imaging.

Mystery of changing choroidal thickness.

Mystery of changing choroidal thickness.