Optical Coherence Tomography Data Research Articles

Assessing the efficacy of 2D and 3D CNN algorithms in OCT-based glaucoma detection

Glaucoma is a progressive neurodegenerative disease characterized by the gradual degeneration of retinal ganglion cells, leading to irreversible blindness worldwide. Therefore, timely and accurate diagnosis of glaucoma is crucial, enabling early intervention and facilitating effective disease management to mitigate further vision deterioration. The advent of optical coherence tomography (OCT) has marked a transformative era in ophthalmology, offering detailed visualization of the macula and optic nerve head (ONH) regions. In recent years, both 2D and 3D convolutional neural network (CNN) algorithms have been applied to OCT image analysis. While 2D CNNs rely on post-prediction aggregation of all B-scans within OCT volumes, 3D CNNs allow for direct glaucoma prediction from the OCT data. However, in the absence of extensively pre-trained 3D models, the comparative efficacy of 2D and 3D-CNN algorithms in detecting glaucoma from volumetric OCT images remains unclear. Therefore, this study explores the efficacy of glaucoma detection through volumetric OCT images using select state-of-the-art (SOTA) 2D-CNN models, 3D adaptations of these 2D-CNN models with specific weight transfer techniques, and a custom 5-layer 3D-CNN-Encoder algorithm. The performance across two distinct datasets is evaluated, each focusing on the macula and the ONH, to provide a comprehensive understanding of the models’ capabilities in identifying glaucoma. Our findings demonstrate that the 2D-CNN algorithm consistently provided robust results compared to their 3D counterparts tested in this study for glaucoma detection, achieving AUC values of 0.960 and 0.943 for the macular and ONH OCT test images, respectively. Given the scarcity of pre-trained 3D models trained on extensive datasets, this comparative analysis underscores the overall utility of 2D and 3D-CNN algorithms in advancing glaucoma diagnostic systems in ophthalmology and highlights the potential of 2D algorithms for volumetric OCT image-based glaucoma detection.

Evaluation of optical coherence tomography findings in patients with multiple sclerosis and glaucoma.

Glaucoma and multiple sclerosis (MS) can cause optic disc pathology and, in this way, affect optical coherence tomography (OCT) data. In this context, the objective of this study is to investigate the changes in the mean, quadrant, and sector data measured by OCT in glaucoma and MS patients. The sample of this prospective cohort study consisted of 42 MS patients (84 eyes), 34 Primary open-angle glaucomas patients (67 eyes), and 24 healthy control subjects (48 eyes). The MS group was divided into two groups according to the presence of a history of optic neuritis. Accordingly, those with a history of optic neuritis were included in the MS ON group, and those without a history of optic neuritis were included in the MS NON group. The differences between these groups in the mean, quadrant, and sector data related to the retinal nerve fiber layer (RNFL) and ganglion cell complex (GCC) were evaluated. Superior nasal (SN), superior temporal (ST), inferior nasal (IN), and superior quadrant (SUP) values were significantly lower in the glaucoma group than in the MS group (p < 0.05). The mean superior GCC (GCC SUP) value was significantly lower in the MS ON group than in the glaucoma group (p < 0.05). On the other hand, SN, ST, inferior temporal (IT), IN, average RNFL (AVE RNFL), semi-average superior RNFL (SUP AVE RNFL), semi-average inferior RNFL (INF AVE RNFL), SUP, and inferior quadrant RNFL (INF) values were significantly lower in the glaucoma group than in the MS NON group (p < 0.05). RNFL and GCC parameters get thinner in MS and glaucoma patients. While the inferior and superior RNFL quadrants are more frequently affected in glaucoma patients, the affected quadrants vary according to the presence of a history of optic neuritis in MS patients. It is noteworthy that the GCC superior quadrant was thin in MS ON patients. The findings of this study indicate that OCT data may be valuable in the differential diagnosis of glaucoma and MS.

Plaque Ruptures Are Related to High Plaque Stress and Strain Conditions: Direct Verification by Using In Vivo OCT Rupture Data and FSI Models.

While it has been hypothesized that high plaque stress and strain may be related to plaque rupture, its direct verification using in vivo coronary plaque rupture data and full 3-dimensional fluid-structure interaction models is lacking in the current literature due to difficulty in obtaining in vivo plaque rupture imaging data from patients with acute coronary syndrome. This case-control study aims to use high-resolution optical coherence tomography-verified in vivo plaque rupture data and 3-dimensional fluid-structure interaction models to seek direct evidence for the high plaque stress/strain hypothesis. In vivo coronary plaque optical coherence tomography data (5 ruptured plaques, 5 no-rupture plaques) were acquired from patients using a protocol approved by the local institutional review board with informed consent obtained. The ruptured caps were reconstructed to their prerupture morphology using neighboring plaque cap and vessel geometries. Optical coherence tomography-based 3-dimensional fluid-structure interaction models were constructed to obtain plaque stress, strain, and flow shear stress data for comparative analysis. The rank-sum test in the nonparametric test was used for statistical analysis. Our results showed that the average maximum cap stress and strain values of ruptured plaques were 142% (457.70 versus 189.22 kPa; P=0.0278) and 48% (0.2267 versus 0.1527 kPa; P=0.0476) higher than that for no-rupture plaques, respectively. The mean values of maximum flow shear stresses for ruptured and no-rupture plaques were 145.02 dyn/cm2 and 81.92 dyn/cm2 (P=0.1111), respectively. However, the flow shear stress difference was not statistically significant. This preliminary case-control study showed that the ruptured plaque group had higher mean maximum stress and strain values. Due to our small study size, larger scale studies are needed to further validate our findings.

A non-invasive capacitive sensor to investigate the Leidenfrost phenomenon: a proof of concept study

A volatile sessile liquid droplet or a sublimating solid manifests levitation on its own vapor when placed on a sufficiently heated surface, illustrating the Leidenfrost phenomenon. In this study, we introduce a non-invasive capacitance method for investigating this phenomenon, offering a potentially simpler alternative to existing optical techniques. The designed sensor features in-plane miniaturized electrodes forming a double-comb structure, also known as an interdigitated capacitor. Initially, the sensor’s capacitance is characterized for various distances between the sensor and a dielectric material. The influence of the sensor substrate material and the spacing between the electrodes on the sensor’s capacitance is also investigated. To demonstrate the feasibility of the method, a sublimating dry ice pellet is placed on the capacitive sensor, and its performance is evaluated. We present results for the dimensionless vapor layer thickness and the pellet’s lifetime at different substrate temperatures, derived from the capacitance output. The results are compared with Optical Coherence Tomography (OCT) data, serving as a benchmark. While the temporal evolution of the sensor’s output, variation in the dimensionless vapor layer thickness, and the lifetime of the dry ice pellet align with expected results from OCT, notable quantitative deviations are observed. These deviations are attributed to practical experimental limitations rather than shortcoming in the sensor’s working principle. Although this necessitates further investigation, the methodology presented in this paper can potentially serve as an alternative for the detection and measurement of Leidenfrost vapor layers.

Unveiling Key Pathological Indicators for Disease Progression in Vogt Koyanagi Harada Disease and Sympathetic Ophthalmia Through Advanced Choroidal Volume Analysis

ABSTRACT Purpose To evaluate the association between quantitative parameters derived from volume analysis of optical coherence tomography (OCT) data and disease worsening in Vogt-Koyanagi-Harada disease (VKHD) and sympathetic ophthalmia (SO). Methods This retrospective study, conducted at Osaka University Hospital, employed swept-source OCT scans from patients diagnosed with VKHD or SO between October 2012 and January 2021. The choroidal vessel structure was segmented and visualized in three dimensions, generating quantitative vessel volume maps. Region-specific choroidal vessel volume (CVV), choroidal volume (CV), and vessel index (VI) were scrutinized for their potential correlation with disease severity. Results Thirty-five eyes of 18 VKHD and 2 SO patient (8 females, 10 males) were evaluated. OCT-derived CVV maps revealed regional CV alterations in VKHD and SO patients. Two parameters, i.e. CV at 3- and 6-month follow-ups (p = 0.044, p = 0.040, respectively, with area under the ROC curve of 0.70) and CVV at 6 months (p = 0.046, area under the ROC curve of 0.71), were significantly higher in recurrent VKHD and SO compared to effectively treated cases. Conclusions The volume analysis of OCT images facilitates a three-dimensional visualization of choroidal alterations, which may serve as a reflection of disease severity in VKHD and SO patients. Furthermore, noninvasive initial CVV or CV measurements may serve as potential biomarkers for predicting disease recurrence in VKHD and SO.

K-Space Approach in Optical Coherence Tomography: Rigorous Digital Transformation of Arbitrary-Shape Beams, Aberration Elimination and Super-Refocusing beyond Conventional Phase Correction Procedures.

For the most popular method of scan formation in Optical Coherence Tomography (OCT) based on plane-parallel scanning of the illuminating beam, we present a compact but rigorous K-space description in which the spectral representation is used to describe both the axial and lateral structure of the illuminating/received OCT signals. Along with the majority of descriptions of OCT-image formation, the discussed approach relies on the basic principle of OCT operation, in which ballistic backscattering of the illuminating light is assumed. This single-scattering assumption is the main limitation, whereas in other aspects, the presented approach is rather general. In particular, it is applicable to arbitrary beam shapes without the need for paraxial approximation or the assumption of Gaussian beams. The main result of this study is the use of the proposed K-space description to analytically derive a filtering function that allows one to digitally transform the initial 3D set of complex-valued OCT data into a desired (target) dataset of a rather general form. An essential feature of the proposed filtering procedures is the utilization of both phase and amplitude transformations, unlike conventionally discussed phase-only transformations. To illustrate the efficiency and generality of the proposed filtering function, the latter is applied to the mutual transformation of non-Gaussian beams and to the digital elimination of arbitrary aberrations at the illuminating/receiving aperture. As another example, in addition to the conventionally discussed digital refocusing enabling depth-independent lateral resolution the same as in the physical focus, we use the derived filtering function to perform digital "super-refocusing." The latter does not yet overcome the diffraction limit but readily enables lateral resolution several times better than in the initial physical focus.

Total variation regularized tensor ring decomposition for OCT image denoising and super-resolution

This paper suggests a novel hybrid tensor-ring (TR) decomposition and first-order tensor-based total variation (FOTTV) model, known as the TRFOTTV model, for super-resolution and noise suppression of optical coherence tomography (OCT) images. OCT imaging faces two fundamental problems undermining correct OCT-based diagnosis: significant noise levels and low sampling rates to speed up the capturing process. Inspired by the effectiveness of TR decomposition in analyzing complicated data structures, we suggest the TRFOTTV model for noise suppression and super-resolution of OCT images. Initially, we extract the nonlocal 3D patches from OCT data and group them to create a third-order low-rank tensor. Subsequently, using TR decomposition, we extract the correlations among all modes of the grouped OCT tensor. Finally, FOTTV is integrated into the TR model to enhance spatial smoothness in OCT images and conserve layer structures more effectively. The proximal alternating minimization and alternative direction method of multipliers are applied to solve the obtained optimization problem. The effectiveness of the suggested method is verified by four OCT datasets, demonstrating superior visual and numerical outcomes compared to state-of-the-art procedures.

Ultra-widefield Fluorescein Angiography and OCT Findings in Children and Young Adults with Autosomal Dominant Neovascular Inflammatory Vitreoretinopathy (ADNIV)

PurposeAutosomal dominant neovascular inflammatory vitreoretinopathy (ADNIV) is a rare genetic (CAPN5) autoimmune condition typically diagnosed in adulthood and characterized by a triad of inflammation, retinal degeneration and neovascularization. We report novel multimodal imaging findings in children and young adults with ADNIV and early treatment response to short-duration local and/or systemic corticosteroids. DesignRetrospective consecutive case series. ParticipantsTen patients under the age of 25 with ADNIV and available multimodal imaging. MethodsThe medical records of patients under the age of 25 with a diagnosis of ADNIV with ultra-widefield fluorescein angiography (UWFFA) and optical coherence tomography (OCT) data were reviewed. Main Outcome MeasuresUWFFA and OCT findings at baseline and following local corticosteroids. ResultsMedian age at presentation was 14 years (range 9-24 years). OCT on presentation demonstrated CME in 8/20 eyes and symptomatic vitreoretinal interface disease in 2/20 eyes. Initial UWFFA demonstrated retinal vascular leakage (20/20 eyes, 100%), peripheral non-perfusion (13/20 eyes, 65%), and retinal neovascularization (6/20 eyes, 30%). Retinal vascular leakage improved with local corticosteroids and neovascularization regressed with anti-VEGF therapy. ConclusionsUWFFA findings of prefibrotic ADNIV reported in adults were also present in children and young adults. Early testing for a pathogenic CAPN5 variant in at-risk children and regularly scheduled screening for uveitis and retinal vasculitis with UWFFA and OCT may prompt earlier intervention. Short duration local steroids are effective at treating retinal vascular leakage and macular edema but are not durable suggesting a potential role for steroid-sparing immunosuppressive therapy. Early treatment may alter the natural history of disease.

Impact of residual stress on coronary plaque stress/strain calculations using optical coherence tomography image-based multi-layer models.

Mechanical stress and strain conditions play an important role in atherosclerosis plaque progression, remodeling and potential rupture and may be used in plaque vulnerability assessment for better clinical diagnosis and treatment decisions. Single layer plaque models without residual stress have been widely used due to unavailability of multi-layer image segmentation method and residual stress data. However, vessel layered structure and residual stress have large impact on stress/strain calculations and should be included in the models. In this study, intravascular optical coherence tomography (OCT) data of coronary plaques from 10 patients were acquired and segmented to obtain the three-layer vessel structure using an in-house automatic segmentation algorithm. Multi- and single-layer 3D thin-slice biomechanical plaque models with and without residual stress were constructed to assess the impact of residual stress on stress/strain calculations. Our results showed that residual stress led to a more uniform stress distribution across the vessel wall, with considerable plaque stress/strain decrease on inner wall and increase on vessel out-wall. Multi-layer model with residual stress inclusion reduced inner wall maximum and mean plaque stresses by 38.57% and 59.70%, and increased out-wall maximum and mean plaque stresses by 572.84% and 432.03%. These findings demonstrated the importance of multi-layer modeling with residual stress for more accurate plaque stress/strain calculations, which will have great impact in plaque cap stress calculation and plaque rupture risk assessment. Further large-scale studies are needed to validate our findings.

Reduced Thickness of the Retina in de novo Parkinson's Disease Shows A Distinct Pattern, Different from Glaucoma.

Parkinson's disease (PD) patients experience visual symptoms and retinal degeneration. Studies using optical coherence tomography (OCT) have shown reduced thickness of the retina in PD, also a key characteristic of glaucoma. To identify the presence and pattern of retinal changes in de novo, treatment-naive PD patients compared to healthy controls (HC) and early primary open angle glaucoma (POAG) patients. Macular OCT data (10×10 mm) were collected from HC, PD, and early POAG patients, at the University Medical Center Groningen. Bayesian informative hypotheses statistical analyses were carried out comparing HC, PD-, and POAG patients, within each retinal cell layer. In total 100 HC, 121 PD, and 78 POAG patients were included. We showed significant reduced thickness of the inner plexiform layer and retinal pigment epithelium in PD compared to HC. POAG patients presented with a significantly thinner retinal nerve fiber layer, ganglion cell layer, inner plexiform layer, outer plexiform layer, and outer photoreceptor and subretinal virtual space compared to PD. Only the outer segment layer and retinal pigment epithelium were significantly thinner in PD compared to POAG. De novo PD patients show reduced thickness of the retina compared to HC, especially of the inner plexiform layer, which differs significantly from POAG, showing a more extensive and widespread pattern of reduced thickness across layers. OCT is a useful tool to detect retinal changes in de novo PD, but its specificity versus other neurodegenerative disorders has to be established.

Mirrored X-Net: Joint classification and contrastive learning for weakly supervised GA segmentation in SD-OCT

Deep learning achieves impressive performance in medical image segmentation, but the training phase requires a large amount of annotated data with precise clinical definitions. Weakly supervised lesion segmentation aims to produce pixel-level masks by learning discriminatory information from weak annotations. In this paper, Mirrored X-Net, a novel segmentation model only supervised by image-level category labels, is proposed to segment the Geographic Atrophy (GA) regions in en-face projection of Spectral-Domain Optical Coherence Tomography (SD-OCT) data. Characterized by the dimension-asymmetric information in SD-OCT images, a novel Anisotropic Downsampling (ADS) is proposed to augment feature shapes. To extract the regions of normal retina for both images with and without lesions, we propose a contrastive learning module to assimilate the deep representation of normal tissue in the SD-OCT images and improve the class-wise difference of deep representation. Based on the contrastive learning module, Anomalous Probability Map (APM) can be obtained to draw the distribution of difference with normal tissue in images. We jointly train the image classification and contrastive learning module, and the final GA segmentation is refined based on the en-face projection of APM. The experimental results on two independent GA datasets demonstrate that the proposed weakly supervised model can produce satisfactory results, and obtain even higher accuracy than fully supervised approaches. The source code is available at https://github.com/maxiao0234/Mirrored-X-Net-pytorch.

Machine learning and optical coherence tomography-derived radiomics analysis to predict persistent diabetic macular edema in patients undergoing anti-VEGF intravitreal therapy

BackgroundDiabetic macular edema (DME) is a leading cause of vision loss in patients with diabetes. This study aimed to develop and evaluate an OCT-omics prediction model for assessing anti-vascular endothelial growth factor (VEGF) treatment response in patients with DME.MethodsA retrospective analysis of 113 eyes from 82 patients with DME was conducted. Comprehensive feature engineering was applied to clinical and optical coherence tomography (OCT) data. Logistic regression, support vector machine (SVM), and backpropagation neural network (BPNN) classifiers were trained using a training set of 79 eyes, and evaluated on a test set of 34 eyes. Clinical implications of the OCT-omics prediction model were assessed by decision curve analysis. Performance metrics (sensitivity, specificity, F1 score, and AUC) were calculated.ResultsThe logistic, SVM, and BPNN classifiers demonstrated robust discriminative abilities in both the training and test sets. In the training set, the logistic classifier achieved a sensitivity of 0.904, specificity of 0.741, F1 score of 0.887, and AUC of 0.910. The SVM classifier showed a sensitivity of 0.923, specificity of 0.667, F1 score of 0.881, and AUC of 0.897. The BPNN classifier exhibited a sensitivity of 0.962, specificity of 0.926, F1 score of 0.962, and AUC of 0.982. Similar discriminative capabilities were maintained in the test set. The OCT-omics scores were significantly higher in the non-persistent DME group than in the persistent DME group (p < 0.001). OCT-omics scores were also positively correlated with the rate of decline in central subfield thickness after treatment (Pearson’s R = 0.44, p < 0.001).ConclusionThe developed OCT-omics model accurately assesses anti-VEGF treatment response in DME patients. The model’s robust performance and clinical implications highlight its utility as a non-invasive tool for personalized treatment prediction and retinal pathology assessment.

Glycemic Variability and the Thickness of Retinal Layers in Cystic Fibrosis Patients with and without Cystic Fibrosis Related Diabetes

Purpose Patients with cystic fibrosis (CF) are at risk to develop CF related diabetes (CFRD) and subsequently even diabetic neuro- and/or vasculopathy. We sought to determine if there are typical signs of diabetes-related retinal alterations present in CF patients with preserved and impaired glycemic control. Methods During routine annual examination CF patients were offered an additional 7-day period of real time continuous glucose monitoring (rtCGM) and an ophthalmological examination including retinal optical coherence tomography (OCT). Patients were categorized according to the glycemic control, i.e. the results of an oral glucose tolerance test (OGTT) and rtCGM were taken into consideration. OCT data was analyzed by our previously published visual analysis software generating dedicated and spatially resolved deviation maps for visualization and quantification of differences in total retinal thickness and thickness of retinal nerve fiber layer (RNFL) as well as ganglion cell layer (GCL) in comparison to age-matched healthy controls and patients with either type 1 or type 2 diabetes mellitus. Results Results of the rtCGM and/or OGTT enabled discrimination between patients with normal glycemic control (CFNG; n = 6), with abnormal glycemic control (CFAG; n = 6) and overt CFRD (n = 4). OCT data indicates gradually increasing retinal thinning in all 3 groups, depending on the degree of glucose metabolism disorder compared to healthy controls. At the foveal region total retinal thickness and GCL thickness were significantly thinner in CFRD patients compared to CFNG patients (total retinal thickness: 260.4 µm (239.3–270.8) vs. 275.4 µm (254.3–289.5); GCL: 11.82 µm (11.16–15.25) vs. 17.30 µm (13.95–19.82); each p < 0.05). Conclusion Although we investigated a rather small number of patients, we obtained evidence that intraretinal neurodegenerative changes occur in each of our subgroups (CFNG, CFAG, CFRD). Beyond this, our results favor the detrimental role of additional diabetes, as the deviations from healthy controls were most pronounced in the CFRD group and are similar to those seen in patients suffering from type 1 or type 2 diabetes.

Evaluation of Increase in Retinal Thickness as Diagnostic Marker in Central Artery Occlusion.

To evaluate the increase in retinal thickness as a marker in predicting the onset of central retinal artery occlusions. Retrospective clinical study conducted at one Swiss hospital. Electronic records were filtered for patients with artery occlusions. Optical coherence tomography data, including time between the imaging and ischemic event, were reviewed. Increase in relative retinal thickness was measured, defined as an increase in retinal thickness compared to the unaffected partner eye. This was correlated with the time from symptom onset. A cutoff value of relative increase of < 24.5% was applied, as suggested in previous studies. The results were compared to the time gathered from the electronic records, and sensitivity, specificity, positive predictive value as well as negative predictive value were calculated for predicting an ischemia time of < 4.5 h. Forty-two eyes from 41 patients with central artery occlusions were identified. Fourteen were female. Mean age was 66.4 ± 15.8 years. Initial corrected visual acuity was 2.41 ± 0.68 logMAR, and 2.13 ± 0.87 logMAR at the last follow-up (p > 0.05). Of eyes with a visual acuity of counting fingers (n = 38) or worse, 89.5% showed no improvement during follow-up, while eyes with logMAR 1 or better (n = 4) improved. Thirteen eyes (13 patients) presented within 4.5 h of the ischemic event. Four patients received i. v. thrombolysis, with visual recovery in one. In 12 eyes with an ischemia time of < 4.5 h, relative increase was below 24.5%. In the remaining 29 eyes with > 4.5 h, relative increase was below 24.5% in 4 eyes and above 24.5% in 25 eyes. This yielded a sensitivity of 92.3%, a specificity of 86.2%, with a positive predictive value of 75.0% and a negative predictive value of 96.2%. Central retinal artery occlusion is associated with severe vision loss. There is no current established therapy. Parameters that objectify the presence of a therapeutic window for thrombolysis are gaining in importance as patient history is often imprecise. Relative retinal thickness increase proved a noninvasive imaging parameter demonstrating adequate performance in detecting patients within the therapeutic window of thrombolysis. Further investigation of this parameter in central retinal occlusion is warranted.

Training in Cortically Blinded Fields Appears to Confer Patient-Specific Benefit Against Retinal Thinning.

Damage to the adult primary visual cortex (V1) causes vision loss in the contralateral hemifield, initiating a process of transsynaptic retrograde degeneration (TRD). Here, we examined retinal correlates of TRD using a new metric to account for global changes in inner retinal thickness and asked if perceptual training in the intact or blind field impacts its progression. We performed a meta-analysis of optical coherence tomography data in 48 participants with unilateral V1 stroke and homonymous visual defects who completed clinical trial NCT03350919. After measuring the thickness of the macular ganglion cell and inner plexiform layer (GCL-IPL) and the peripapillary retinal nerve fiber layer (RNFL), we computed individual laterality indices (LI) at baseline and after ∼6months of daily motion discrimination training in the intact or blind field. Increasingly positive LI denoted greater layer thinning in retinal regions affected versus unaffected by the cortical damage. Pretraining, the affected GCL-IPL and RNFL were thinner than their unaffected counterparts, generating LI values positively correlated with time since stroke. Participants trained in their intact field exhibited increased LIGCL-IPL. Those trained in their blind field had no significant change in LIGCL-IPL. LIRNFL did not change in either group. Relative shrinkage of the affected versus unaffected macular GCL-IPL can be reliably measured at an individual level and increases with time post-V1 stroke. Relative thinning progressed during intact-field training but appeared to be halted by training within the blind field, suggesting a potentially neuroprotective effect of this simple behavioral intervention.

Accuracy of generative deep learning model for macular anatomy prediction from optical coherence tomography images in macular hole surgery

This study aims to propose a generative deep learning model (GDLM) based on a variational autoencoder that predicts macular optical coherence tomography (OCT) images following full-thickness macular hole (FTMH) surgery and evaluate its clinical accuracy. Preoperative and 6-month postoperative swept-source OCT data were collected from 150 patients with successfully closed FTMH using 6 × 6 mm2 macular volume scan datasets. Randomly selected and augmented 120,000 training and 5000 validation pairs of OCT images were used to train the GDLM. We assessed the accuracy and F1 score of concordance for neurosensory retinal areas, performed Bland–Altman analysis of foveolar height (FH) and mean foveal thickness (MFT), and predicted postoperative external limiting membrane (ELM) and ellipsoid zone (EZ) restoration accuracy between artificial intelligence (AI)-OCT and ground truth (GT)-OCT images. Accuracy and F1 scores were 94.7% and 0.891, respectively. Average FH (228.2 vs. 233.4 μm, P = 0.587) and MFT (271.4 vs. 273.3 μm, P = 0.819) were similar between AI- and GT-OCT images, within 30.0% differences of 95% limits of agreement. ELM and EZ recovery prediction accuracy was 88.0% and 92.0%, respectively. The proposed GDLM accurately predicted macular OCT images following FTMH surgery, aiding patient and surgeon understanding of postoperative macular features.

Two-stage adversarial learning based unsupervised domain adaptation for retinal OCT segmentation.

Deep learning based optical coherence tomography (OCT) segmentation methods have achieved excellent results, allowing quantitative analysis of large-scale data. However, OCT images are often acquired by different devices or under different imaging protocols, which leads to serious domain shift problem. This in turn results in performance degradation of segmentation models. Aiming at the domain shift problem, we propose a two-stage adversarial learning based network (TSANet) that accomplishes unsupervised cross-domain OCT segmentation. In the first stage, a Fourier transform based approach is adopted to reduce image style differences from the image level. Then, adversarial learning networks, including a segmenter and a discriminator, are designed to achieve inter-domain consistency in the segmentation output. In the second stage, pseudo labels of selected unlabeled target domain training data are used to fine-tune the segmenter, which further improves its generalization capability. The proposed method was tested on cross-domain datasets for choroid or retinoschisis segmentation tasks. For choroid segmentation, the model was trained on 400 images and validated on 100 images from the source domain, and then trained on 1320 unlabeled images and tested on 330 images from target domain I, and also trained on 400 unlabeled images and tested on 200 images from target domain II. For retinoschisis segmentation, the model was trained on 1284 images and validated on 312 images from the source domain, and then trained on 1024 unlabeled images and tested on 200 images from the target domain. The proposed method achieved significantly improved results over that without domain adaptation, with improvement of 8.34%, 55.82% and 3.53% in intersection over union (IoU) respectively for the three test sets. The performance is better than some state-of-the-art domain adaptation methods. The proposed TSANet, with image level adaptation, feature level adaptation and pseudo-label based fine-tuning, achieved excellent cross-domain generalization. This alleviates the burden of obtaining additional manual labels when adapting the deep learning model to new OCT data.

Short-term outcomes of treatment switch to faricimab in patients with aflibercept-resistant neovascular age-related macular degeneration.

To report short-term outcomes of treatment switch to faricimab in real-world patients with aflibercept-resistant neovascular age-related macular degeneration (AMD). Single-center, retrospective cohort study with chart-review using electronic injection database, electronic medical records, and optical coherence tomography (OCT) data from May to September 2023. A total of 50 eyes of 46 patients were analyzed. Faricimab treatment led to absence of fluid in 32% of the eyes and a reduction of fluid in 84% of the eyes. There was a statistically significant decrease in central retinal thickness (CRT) and pigment epithelial detachment (PED) height in those that responded to the switch (median difference: - 31μm, IQR: 55, p < 0.0001 and median difference: - 21μm, IQR: 36, p < 0.0001, respectively) and a statistically significant increase in CRT (median difference: + 19μm, IQR: 20, p = 0.0143) and no change in PED height (median difference: + 22μm, IQR: 64, p = 0.1508) in those that did not. Best-corrected visual acuity (BCVA) showed marginal decrease with low statistical significance. No ocular or systemic safety events were observed. Our findings suggest that switching to faricimab is generally safe and effective in patients with neovascular AMD who are otherwise difficult to treat and have residual fluid despite frequent injections with aflibercept. We observed a high rate of morphological response to the treatment switch, improvement of anatomical parameters with about one-third of patients having dry macula following a single injection, and a marginal change in BCVA. Sustainability of these results requires further investigation. ClinicalTrials.gov registration number: NCT06124677. Date of registration: 09/11/2023, retrospectively registered.

Japan-epiretinal membrane (J-ERM) registry: A prospective cohort study protocol investigating the surgical outcome of epiretinal membrane.

Epiretinal membrane (ERM) causes visual impairment such as reduction in visual acuity and metamorphopsia due to retinal traction. With the improvement of optical coherence tomography (OCT) and microincision vitrectomy surgery (MIVS), the surgery of ERM has significantly advanced. However, there have been no large-scale studies on the following: (1) how to evaluate visual impairment in ERM, (2) the relationship between OCT findings and visual function, (3) when is the optimal timing of surgery, and (4) the relationship between the surgical instruments as well as techniques and prognosis. The purpose of this study was to obtain evidence regarding these ERM surgeries. This is a prospective, multicenter cohort study of ERM surgery in Japan from March 1, 2023, to March 31, 2027 (UMIN000048472, R-3468-2). Patients who underwent ERM surgery during the study period and agreed to participate in this study will be included. The goal is to have a total of 5,000 eyes surgically treated for ERM. The following data will be collected: age, gender, medical history, subjective symptoms, visual function before and 6 and 12 months after surgery, clinical findings, OCT data, surgical technique, instruments used in surgery, and complications. The results of this study will support the surgical decisions and procedures in ERM practices.

Comparative results of use and switching of anti-TNF-alpha agents in adult Behçet's uveitis.

This study was planned to compare the clinical results and optical coherence tomography (OCT) data of patients who underwent adalimumab (ADA), infliximab (IFX) and switched protocols between these two methods in cases of Behçet's associated uveitis (BAU). The patients with BAU who were treated with anti-TNF- α agents for at least one year were qualified for the chart review. The outcome parameters were the steroid-free remission state, the best-corrected visual acuity (BCVA), the central macular thickness (CMT), and the subfoveal choroidal thickness (SFCT) of the patients at the last visit. After a mean follow-up of 16.60 months, the rate of active inflammation decreased from 63.6% to 15.1%. Mean BCVA increased, mean CMT and mean SFCT decreased significantly at the last visit ( P < 0.001, P < 0.001, respectively). Anti-TNF- α treatment and switching protocols between these two agents effectively and safely control intraocular inflammation in BAU patients.