Deep learning models for screening of high myopia using optical coherence tomography
Ontology highlight
ABSTRACT:
SUBMITTER: Choi K
PROVIDER: S-EPMC8568935 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:PurposeThe purpose of this study was to engineer deep learning (DL) models that can identify myopic maculopathy in patients with high myopia based on optical coherence tomography (OCT) images.MethodsAn artificial intelligence (AI) system was developed using 2342 qualified OCT macular images from 1041 patients with pathologic myopia admitted to the Affiliated Eye Hospital of Wenzhou Medical University (WMU). We adopted an ResNeSt101 architecture to train five independent models to identify the following five myopic maculopathies: macular choroidal thinning, macular Bruch membrane (BM) defects, subretinal hyper-reflective material (SHRM), myopic traction maculopathy (MTM), and dome-shaped macula (DSM). We tested the models with an independent test dataset that included 450 images obtained from 297 patients with high myopia. Focal loss was used to address class imbalance, and optimal operating thresholds were determined according to the Youden Index. The performance was quantified using the area under the receiver operating characteristic (AUC), sensitivity, specificity, and confusion matrix.ResultsFor the identification of myopic maculopathy, the AUCs of receiver operating characteristic (ROC) curves were 0.927 to 0.974 for 5 myopic maculopathies. Our AI system achieved sensitivities equal to or even better than those of junior retinal specialists (56.16-99.73%). The diagnosis of it is also interpretable that we provide visual explanations clearly via heatmaps.ConclusionsWe developed a convolutional neural network (CNN)-based DL AI system for detection and classification of myopic maculopathy in patients with high myopia using OCT macular images. Our AI system achieved sensitivities equal to or even better than those of junior retinal specialists.Translational relevanceThis AI system can be widely applied in sophisticated situations in large-scale high myopia screening.
| S-EPMC8590175 | biostudies-literature
Project description:Several studies published so far used highly selective image datasets from unclear sources to train computer vision models and that may lead to overestimated results, while those studies conducted in real-life remain scarce. To avoid image selection bias, we stacked convolutional and recurrent neural networks (CNN-RNN) to analyze complete optical coherence tomography (OCT) cubes in a row and predict diabetic macular edema (DME), in a real-world diabetic retinopathy screening program. A retrospective cohort study was carried out. Throughout 4-years, 5314 OCT cubes from 4408 subjects who attended to the diabetic retinopathy (DR) screening program were included. We arranged twenty-two (22) pre-trained CNNs in parallel with a bidirectional RNN layer stacked at the bottom, allowing the model to make a prediction for the whole OCT cube. The staff of retina experts built a ground truth of DME later used to train a set of these CNN-RNN models with different configurations. For each trained CNN-RNN model, we performed threshold tuning to find the optimal cut-off point for binary classification of DME. Finally, the best models were selected according to sensitivity, specificity, and area under the receiver operating characteristics curve (AUROC) with their 95% confidence intervals (95%CI). An ensemble of the best models was also explored. 5188 cubes were non-DME and 126 were DME. Three models achieved an AUROC of 0.94. Among these, sensitivity, and specificity (95%CI) ranged from 84.1-90.5 and 89.7-93.3, respectively, at threshold 1, from 89.7-92.1 and 80-83.1 at threshold 2, and from 80.2-81 and 93.8-97, at threshold 3. The ensemble model improved these results, and lower specificity was observed among subjects with sight-threatening DR. Analysis by age, gender, or grade of DME did not vary the performance of the models. CNN-RNN models showed high diagnostic accuracy for detecting DME in a real-world setting. This engine allowed us to detect extra-foveal DMEs commonly overlooked in other studies, and showed potential for application as the first filter of non-referable patients in an outpatient center within a population-based DR screening program, otherwise ended up in specialized care.
| S-EPMC11291805 | biostudies-literature
Project description:The choroid layer is a vascular layer in human retina and its main function is to provide oxygen and support to the retina. Various studies have shown that the thickness of the choroid layer is correlated with the diagnosis of several ophthalmic diseases. For example, diabetic macular edema (DME) is a leading cause of vision loss in patients with diabetes. Despite contemporary advances, automatic segmentation of the choroid layer remains a challenging task due to low contrast, inhomogeneous intensity, inconsistent texture and ambiguous boundaries between the choroid and sclera in Optical Coherence Tomography (OCT) images. The majority of currently implemented methods manually or semi-automatically segment out the region of interest. While many fully automatic methods exist in the context of choroid layer segmentation, more effective and accurate automatic methods are required in order to employ these methods in the clinical sector. This paper proposed and implemented an automatic method for choroid layer segmentation in OCT images using deep learning and a series of morphological operations. The aim of this research was to segment out Bruch's Membrane (BM) and choroid layer to calculate the thickness map. BM was segmented using a series of morphological operations, whereas the choroid layer was segmented using a deep learning approach as more image statistics were required to segment accurately. Several evaluation metrics were used to test and compare the proposed method against other existing methodologies. Experimental results showed that the proposed method greatly reduced the error rate when compared with the other state-of-the-art methods.
| S-EPMC6395677 | biostudies-literature
Project description:We developed a deep learning architecture based on Inception V3 to predict visual field using optical coherence tomography (OCT) imaging and evaluated its performance. Two OCT images, macular ganglion cell-inner plexiform layer (mGCIPL) and peripapillary retinal nerve fibre layer (pRNFL) thicknesses, were acquired and combined. A convolutional neural network architecture was constructed to predict visual field using this combined OCT image. The root mean square error (RMSE) between the actual and predicted visual fields was calculated to evaluate the performance. Globally (the entire visual field area), the RMSE for all patients was 4.79 ± 2.56 dB, with 3.27 dB and 5.27 dB for the normal and glaucoma groups, respectively. The RMSE of the macular region (4.40 dB) was higher than that of the peripheral region (4.29 dB) for all subjects. In normal subjects, the RMSE of the macular region (2.45 dB) was significantly lower than that of the peripheral region (3.11 dB), whereas in glaucoma subjects, the RMSE was higher (5.62 dB versus 5.03 dB, respectively). The deep learning method effectively predicted the visual field 24-2 using the combined OCT image. This method may help clinicians determine visual fields, particularly for patients who are unable to undergo a physical visual field exam.
| S-EPMC7337305 | biostudies-literature
Project description:Polarization-sensitive optical coherence tomography (PS-OCT) is a high-resolution label-free optical biomedical imaging modality that is sensitive to the microstructural architecture in tissue that gives rise to form birefringence, such as collagen or muscle fibers. To enable polarization sensitivity in an OCT system, however, requires additional hardware and complexity. We developed a deep-learning method to synthesize PS-OCT images by training a generative adversarial network (GAN) on OCT intensity and PS-OCT images. The synthesis accuracy was first evaluated by the structural similarity index (SSIM) between the synthetic and real PS-OCT images. Furthermore, the effectiveness of the computational PS-OCT images was validated by separately training two image classifiers using the real and synthetic PS-OCT images for cancer/normal classification. The similar classification results of the two trained classifiers demonstrate that the predicted PS-OCT images can be potentially used interchangeably in cancer diagnosis applications. In addition, we applied the trained GAN models on OCT images collected from a separate OCT imaging system, and the synthetic PS-OCT images correlate well with the real PS-OCT image collected from the same sample sites using the PS-OCT imaging system. This computational PS-OCT imaging method has the potential to reduce the cost, complexity, and need for hardware-based PS-OCT imaging systems.
| S-EPMC8249385 | biostudies-literature
Project description:As the prevalence of diabetes increases, millions of people need to be screened for diabetic retinopathy (DR). Remarkable advances in technology have made it possible to use artificial intelligence to screen DR from retinal images with high accuracy and reliability, resulting in reducing human labor by processing large amounts of data in a shorter time. We developed a fully automated classification algorithm to diagnose DR and identify referable status using optical coherence tomography angiography (OCTA) images with convolutional neural network (CNN) model and verified its feasibility by comparing its performance with that of conventional machine learning model. Ground truths for classifications were made based on ultra-widefield fluorescein angiography to increase the accuracy of data annotation. The proposed CNN classifier achieved an accuracy of 91-98%, a sensitivity of 86-97%, a specificity of 94-99%, and an area under the curve of 0.919-0.976. In the external validation, overall similar performances were also achieved. The results were similar regardless of the size and depth of the OCTA images, indicating that DR could be satisfactorily classified even with images comprising narrow area of the macular region and a single image slab of retina. The CNN-based classification using OCTA is expected to create a novel diagnostic workflow for DR detection and referral.
| S-EPMC8626435 | biostudies-literature
Project description:PurposeTo develop an automated diabetic retinopathy (DR) staging system using optical coherence tomography angiography (OCTA) images with a convolutional neural network (CNN) and to verify the feasibility of the system.MethodsIn this retrospective cross-sectional study, a total of 918 data sets of 3 × 3 mm2 OCTA images and 917 data sets of 6 × 6 mm2 OCTA images were obtained from 1118 eyes. A deep CNN and four traditional machine learning models were trained with annotations made by a retinal specialist based on ultra-widefield fluorescein angiography. Separately, the same images of the test data sets were independently graded by two human experts. The results of the CNN algorithm were compared with those of traditional machine learning-based classifiers and human experts.ResultsThe proposed CNN achieved an accuracy of 0.728, a sensitivity of 0.675, a specificity of 0.944, an F1 score of 0.683, and a quadratic weighted κ of 0.908 for a six-level staging task, which were far superior to the results of traditional machine learning methods or human experts. The CNN algorithm showed a better performance using 6 × 6 mm2 rather than 3 × 3 mm2 sized OCTA images and using combined data rather than a separate OCTA layer alone.ConclusionsCNN-based classification using OCTA images can provide reliable assistance to clinicians for DR classification.Translational relevanceThis CNN algorithm can guide the clinical decision for invasive angiography or referrals to ophthalmology specialists, helping to create more efficient diagnostic workflow in primary care settings.
| S-EPMC8899862 | biostudies-literature
Project description:PurposeTo develop deep learning (DL) models to predict best-corrected visual acuity (BCVA) from optical coherence tomography (OCT) images from patients with neovascular age-related macular degeneration (nAMD).MethodsRetrospective analysis of OCT images and associated BCVA measurements from the phase 3 HARBOR trial (NCT00891735). DL regression models were developed to predict BCVA at the concurrent visit and 12 months from baseline using OCT images. Binary classification models were developed to predict BCVA of Snellen equivalent of <20/40, <20/60, and ≤20/200 at the concurrent visit and 12 months from baseline.ResultsThe regression model to predict BCVA at the concurrent visit had R 2 = 0.67 (root-mean-square error [RMSE] = 8.60) in study eyes and R 2 = 0.84 (RMSE = 9.01) in fellow eyes. The best classification model to predict BCVA at the concurrent visit had an area under the receiver operating characteristic curve (AUC) of 0.92 in study eyes and 0.98 in fellow eyes. The regression model to predict BCVA at month 12 using baseline OCT had R 2 = 0.33 (RMSE = 14.16) in study eyes and R 2 = 0.75 (RMSE = 11.27) in fellow eyes. The best classification model to predict BCVA at month 12 had AUC = 0.84 in study eyes and AUC = 0.96 in fellow eyes.ConclusionsDL shows promise in predicting BCVA from OCTs in nAMD. Further research should elucidate the utility of models in clinical settings.Translational relevanceDL models predicting BCVA could be used to enhance understanding of structure-function relationships and develop more efficient clinical trials.
| S-EPMC7488630 | biostudies-literature
Project description:To compare four different optical coherence tomography (OCT) devices for visualization of retinal and subretinal layers in highly myopic eyes. In this prospective, observational, cross-sectional study, consecutive patients with high myopia and control subjects were imaged by four OCT devices: Spectralis OCT2, PlexElite 2.0 100 kHz, PlexElite 2.0 200 kHz and the Canon Xephilio OCT-S1. The acquisition protocol for comparison consisted of single vertical and horizontal line scans centered on the fovea. Comparison between the devices in the extent of visible retina, presence of conjugate image or mirror artifacts, visibility of the sclerochoroidal interface and retrobulbar tissue. 30 eyes with high myopia and 30 control subjects were analyzed. The visualized RPE length was significantly different between the OCT devices with Xephilio OCT-S1 imaging the largest extent (p < 0.0001). The proportion of eyes with conjugate image artifact was significantly higher with the Spectralis OCT (p < 0.0001), and lower with the PlexElite 200 kHz (p < 0.0001). No difference in visibility of the sclerochoroidal interface was noted among instruments. The retrobulbar tissue was visible in a higher proportion of eyes using swept-source PlexElite 100 kHz and 200 kHz (p < 0.007) compared to the other devices. In highly myopic eyes, the four OCT devices demonstrated significant differences in the extent of the retina imaged, in the prevalence of conjugate image artifact, and in the visualization of the retrobulbar tissue.
| S-EPMC8067388 | biostudies-literature
Project description:BackgroundWhile deep learning has delivered promising results in the field of ophthalmology, the hurdle to complete a deep learning study is high. In this study, we aim to facilitate small scale model trainings by exploring the role of preprocessing to reduce computational burden and accelerate learning.MethodsA small subset of a previously published dataset containing optical coherence tomography images of choroidal neovascularization, drusen, diabetic macula edema, and normal macula was modified using Fourier transformation and bandpass filter, producing high frequency images, original images, and low frequency images. Each set of images was trained with the same model, and their performances were compared.ResultsCompared to that with the original image dataset, the model trained with the high frequency image dataset achieved an improved final performance and reached maximum performance much earlier (in fewer epochs). The model trained with low frequency images did not achieve a meaningful performance.ConclusionAppropriate preprocessing of training images can accelerate the training process and can potentially facilitate modeling using artificial intelligence when limited by sample size or computational power.
| S-EPMC10108538 | biostudies-literature