Optical Coherence Tomography B-scans Research Articles

Enhanced microvascular imaging through deep learning-driven OCTA reconstruction with squeeze-and-excitation block integration

Skin microvasculature is essential for cardiovascular health and thermoregulation in humans, yet its imaging and analysis pose significant challenges. Established methods, such as speckle decorrelation applied to optical coherence tomography (OCT) B-scans for OCT-angiography (OCTA), often require a high number of B-scans, leading to long acquisition times that are prone to motion artifacts. In our study, we propose a novel approach integrating a deep learning algorithm within our OCTA processing. By integrating a convolutional neural network with a squeeze-and-excitation block, we address these challenges in microvascular imaging. Our method enhances accuracy and reduces measurement time by efficiently utilizing local information. The Squeeze-and-Excitation block further improves stability and accuracy by dynamically recalibrating features, highlighting the advantages of deep learning in this domain.

Quantitative analysis of peripapillary capillary volume using dense B-scan OCT angiography in normal and diabetic retina

BackgroundThe value of quantitatively analyzing peripapillary capillary volume (PPCV) distribution was explored in normal and diabetic retinopathy (DR) eyes using dense B-scan optical coherence tomography angiography (DB OCTA).MethodsThis was a cross-sectional observational study followed by prospective follow-up for those with DR, which enrolled 101 healthy subjects and 140 DR patients. Dense, automatic, real-time (DART) volume scans of DB OCTA were performed using a Spectralis HRA + OCT2. ImageJ and MATLAB were used to process and calculate PPCV distribution detected by DB OCTA.ResultsIn normal subjects, PPCV distribution were significantly correlated with the age and quadrant location (all P < 0.001). The PPCV distribution in each quadrant was significantly lower in severe nonproliferative DR patients than in normal subjects in all age groups (all P < 0.05, t-test). Compared to normal subjects, the PPCV distribution improved significantly in the pan-retinal photocoagulation treatment and surgery groups (all P < 0.001). No significant variation was observed in the anti-VEGF treatment group and normal subjects (P > 0.05). The PPCV distribution is significantly correlated with post-treatment best-corrected visual acuity in both the pan-retinal photocoagulation treatment and surgery groups (all P < 0.003) but not in the anti-VEGF treatment group (P = 0.940).ConclusionsQuantitative assessment of PPCV distribution using DB OCTA is valuable in prognosis evaluation of DR with pan-retinal photocoagulation and surgery.

Self-supervised contrastive learning improves machine learning discrimination of full thickness macular holes from epiretinal membranes in retinal OCT scans.

There is a growing interest in using computer-assisted models for the detection of macular conditions using optical coherence tomography (OCT) data. As the quantity of clinical scan data of specific conditions is limited, these models are typically developed by fine-tuning a generalized network to classify specific macular conditions of interest. Full thickness macular holes (FTMH) present a condition requiring urgent surgical repair to prevent vision loss. Other works on automated FTMH classification have tended to use supervised ImageNet pre-trained networks with good results but leave room for improvement. In this paper, we develop a model for FTMH classification using OCT B-scans around the central foveal region to pre-train a naïve network using contrastive self-supervised learning. We found that self-supervised pre-trained networks outperform ImageNet pre-trained networks despite a small training set size (284 eyes total, 51 FTMH+ eyes, 3 B-scans from each eye). On three replicate data splits, 3D spatial contrast pre-training yields a model with an average F1-score of 1.0 on holdout data (50 eyes total, 10 FTMH+), compared to an average F1-score of 0.831 for FTMH detection by ImageNet pre-trained models. These results demonstrate that even limited data may be applied toward self-supervised pre-training to substantially improve performance for FTMH classification, indicating applicability toward other OCT-based problems.

Management of Perifoveal Exudative Vascular Anomalous Complex.

To report a case of a patient with perifoveal exudative vascular anomalous complex (PEVAC) and to analyze morphological and vascular changes by OCT and OCTA before and after treatment. In this case we reported a case which received multiple treatment for unresponsive effect to Repeated Aflibercept intravitreal injections, subthreshold micropulse laser therapy (SMPL) duty cycle 5%. At last the focal fully dose yellow laser was performed. A 57-year-old male patient presented with best corrected visual acuity (BCVA) of 20/50 in his right eye (RE) The fundoscopic evaluation, structural optical coherence tomography (OCT) and OCT angiography (OCTA) deposed for a diagnosis of PEVAC. The patient underwent three Aflibercept intravitreal injections in the RE. Since no changes were detected, we opted for multiple sessions of yellow subthreshold micropulse laser therapy (SMPL) duty cycle 5%, however no signs of regression were registered. Only after a fully dose yellow laser, signs of edema regression were observed. Stuctural OCT B-scan showed complete restitutio of retinal profile, in absence of any intraretinal or subretinal fluid and PEVAC lesion, while the OCTA showed a slight flow deficit at the previous lesion site. and Importance: The PEVAC treatment is still unknown. In our experience, we reported a case of multiple treatments for unresponsive effect to: Aflibercept intravitreal injections, subthreshold micropulse laser therapy (SMPL) duty cycle 5%. At list the focal fully dose yellow laser was the only effective in our patient. We proposed our management to share the heterogeneous response at PEVAC entity.

Integrated line-field optical coherence tomography and scheimpflug imaging for corneal imaging

Previously we demonstrated a combined scanning-point optical coherence tomography (OCT) and Scheimpflug Imaging (SI) system, to provide cross-section images of the corneal layers and the whole anterior segment of the eye simultaneously. However, to fully realise the benefits of this dual imaging modality concept towards clinical practice, it is required that the OCT part of the system can achieve ultra-high axial resolution (<3 µm in air) and faster imaging speeds, without using prohibitively expensive components. To achieve this, a new imaging device integrating line-field OCT and SI, using a supercontinuum light source, is demonstrated. This line-field configuration enables both OCT and SI B-scan to be captured using the same light illumination in a single shot, which is the most significant improvement over single point scanning. This achieves an OCT axial resolution down to 2.1 µm in air (1.6 µm in corneal tissue) and an imaging speed up to 213 kA-Scans/s. Since the OCT and SI capture the identical corneal position and serve as cross-validation, this technique offers a great method for accurately determining the individual corneal thickness and refractive index, thereby minimizing individual variations, which can help determine the extent of cutting or correction needed in vision correction. Moreover, the precise measurement of corneal thickness provided by this technique allows for a better understanding of the biomechanical properties of the cornea. We demonstrate the efficacy of the proposed system by evaluating porcine and bovine eyes ex-vivo for determining the corneal refractive index and thickness.

Prevalence and significance of fovea plana in patients with epiretinal membrane.

Fovea plana is indicative of an immature inner retina, yet its association with epimacular membrane (ERM) remains unexplored. This study aims to investigate the prevalence of fovea plana in the contralateral eye of patients diagnosed with ERM. A retrospective analysis was conducted on consecutive patients in a monocentric tertiary ophthalmic department between January and April 2021. The study included all patients referred for ERM, with dense optical coherence tomography (OCT) B-scans utilized to assess the incidence of fovea plana in the contralateral eye. Patients with bilateral ERM were excluded from the analysis. Out of 181 patients, 26 exhibited fovea plana in the contralateral eye, resulting in an incidence rate of 14.3%. Demographic characteristics, OCT patterns, and indications for surgery did not differ significantly between patients with and without fovea plana. The prevalence of fovea plana in patients with ERM did not exhibit a statistically significant increase compared to the reported incidence in patients undergoing preoperative cataract surgery. In our cohort, the incidence of fovea plana in patients with ERM was determined to be 14.3%.

Application of novel non-invasive ophthalmic imaging to visualize peripapillary wrinkles, retinal folds and peripapillary hyperreflective ovoid mass-like structures associated with elevated intracranial pressure.

Elevated intracranial pressure (ICP) is a serious and potentially life-threatening condition, for which clinically useful non-invasive measures have been elusive, in some cases due to their inadequate sensitivity and specificity. Our aim was to evaluate novel non-invasive ophthalmic imaging of selected pathological features seen in elevated ICP, namely peripapillary hyperreflective ovoid mass-like structures (PHOMS), peripapillary wrinkles (PPW) and retinal folds (RF) as potential biomarkers of elevated ICP. This single-center pilot study included subjects with untreated or incompletely treated high ICP. The retinas of these subjects were evaluated with averaged en-face optical coherence tomography (OCT), OCT retinal cross-sections (OCT B-scans), adaptive optics scanning light ophthalmoscopy (AOSLO), and fundus photos. Seven subjects were included in the study. 6 subjects with high ICP (5 idiopathic intracranial hypertension, 1 medication induced, 30.8 ± 8.6 years, 75% female, 5 with papilledema) and 1 control (20-25 years) were included. PHOMS, PPW and RF were present in all subjects with papilledema, but neither in the high ICP subject without papilledema nor in the control subject. Averaged en-face OCT scans and AOSLO were more sensitive for PPW and RF than OCT B-scans and commercial fundus photos. PPW, RF and PHOMS volume have potential as non-invasive biomarkers of ICP. Novel imaging modalities may improve sensitivity. However, lack of automated image acquisition and processing limits current widespread adoption in clinical settings. Further research is needed to validate these structures as biomarkers for elevated ICP and improve clinical utility.

Time-Resolved Dynamic Optical Coherence Tomography for Retinal Blood Flow Analysis.

Optical coherence tomography (OCT) representations in clinical practice are static and do not allow for a dynamic visualization and quantification of blood flow. This study aims to present a method to analyze retinal blood flow dynamics using time-resolved structural OCT. We developed novel imaging protocols to acquire video-rate time-resolved OCT B-scans (1024 × 496 pixels, 10 degrees field of view) at four different sensor integration times (integration time of 44.8 µs at a nominal A-scan rate of 20 kHz, 22.4 µs at 40 kHz, 11.2 µs at 85 kHz, and 7.24 µs at 125 kHz). The vessel centers were manually annotated for each B-scan and surrounding subvolumes were extracted. We used a velocity model based on signal-to-noise ratio (SNR) drops due to fringe washout to calculate blood flow velocity profiles in vessels within five optic disc diameters of the optic disc rim. Time-resolved dynamic structural OCT revealed pulsatile SNR changes in the analyzed vessels and allowed the calculation of potential blood flow velocities at all integration times. Fringe washout was stronger in acquisitions with longer integration times; however, the ratio of the average SNR to the peak SNR inside the vessel was similar across all integration times. We demonstrated the feasibility of estimating blood flow profiles based on fringe washout analysis, showing pulsatile dynamics in vessels close to the optic nerve head using structural OCT. Time-resolved dynamic OCT has the potential to uncover valuable blood flow information in clinical settings.

Deep Learning Versus Corneal Tomography Features to Detect Subclinical Corneal Edema in Fuchs Endothelial Corneal Dystrophy

Purpose: To compare a deep learning model with corneal tomography features for detecting subclinical corneal edema in patients with Fuchs endothelial corneal dystrophy (FECD). Methods: We trained a deep learning model to detect corneal edema on 379 optical coherence tomography B-scans of normal and edematous corneas. 51 eyes of 32 patients with FECD were analyzed and compared with 100 eyes of 50 normal patients. For each eye, the cornea was scanned on the same day using 2 modalities of the same swept-source optical coherence tomography device (Anterion): corneal tomography maps and 6 high-resolution radial B-scans. The 6 radial B-scans were analyzed using our model from which an en face map of edema was reconstructed. The location exhibiting the highest probability of edema was derived from that map. Two corneal surgeons assessed the tomography maps and labeled the location of the supposed highest edema. This location was compared with our model's en face map. Results: According to tomography features, 64.7% of eyes presented subclinical edema. Our model and tomography features agreed in 80% of cases for the presence or absence of subclinical edema. The average distance between the location of maximal edema determined by human experts on tomography maps and that derived from our model's en face map was 1.91 ± 1.15 mm. Our model detected superior peripheral edema in a majority of eyes. Conclusions: Despite being based on different approaches, both methods agreed in the detection of subclinical edema in most cases. The location of detected edema was very similar in both methods. In cases where both methods disagree, our approach provides new objective results that might help the surgeon in making a decision in difficult cases.

Progression or Aging? A Deep Learning Approach for Distinguishing Glaucoma Progression From Age-Related Changes in OCT Scans

To develop deep learning (DL) algorithm to detect glaucoma progression using optical coherence tomography (OCT) images, in the absence of a reference standard. Retrospective cohort study. Glaucomatous and healthy eyes with ≥5 reliable peripapillary OCT (Spectralis, Heidelberg Engineering) circle scans were included. A weakly supervised time-series learning model, called noise positive-unlabeled (Noise-PU) DL was developed to classify whether sequences of OCT B-scans showed glaucoma progression. The model used 2 learning schemes, one to identify age-related changes by differentiating test sequences from glaucoma vs healthy eyes, and the other to identify test-retest variability based on scrambled OCTs of glaucoma eyes. Both models' bases were convolutional neural networks (CNN) and long short-term memory (LSTM) networks which were combined to form a CNN-LSTM model. Model features were combined and jointly trained to identify glaucoma progression, accounting for age-related loss. The DL model's outcomes were compared with ordinary least squares (OLS) regression of retinal nerve fiber layer (RNFL) thickness over time, matched for specificity. The hit ratio was used as a proxy for sensitivity. Eight thousand seven hundred eighty-five follow-up sequences of 5 consecutive OCT tests from 3253 eyes (1859 subjects) were included in the study. The mean follow-up time was 3.5 ± 1.6 years. In the test sample, the hit ratios of the DL and OLS methods were 0.498 (95%CI: 0.470-0.526) and 0.284 (95%CI: 0.258-0.309) respectively (P < .001) when the specificities were equalized to 95%. A DL model was able to identify longitudinal glaucomatous structural changes in OCT B-scans using a surrogate reference standard for progression.

Vitreoretinal interface abnormalities in fellow eyes of patients with a macular hole.

This study aimed to determine the incidence of bilateral macular holes (MHs), and to investigate the incidence and characteristics of retinal abnormalities in the fellow eyes, using high-definition optical coherence tomography (OCT). Medical records from 724 consecutive patients diagnosed with full-thickness MHs, who underwent surgery at Fujita Health University Hospital from May 2015 to March 2022, were retrospectively reviewed. Based on the exclusion criteria, 658 patients with idiopathic MHs were included. B-scan OCT images of the fellow eyes were used to identify vitreoretinal interface abnormalities, such as MHs, epiretinal membranes (ERMs), vitreomacular traction syndrome (VMTS) and lamellar macular holes (LMHs). The mean age of the patients was 65.9±9.0 years. There were 292 males (44.4%) and 366 females (55.6%). There were 39 fellow eyes with MHs (5.9%), 77 with ERMs (11.7%), 32 with VMTS (4.9%) and 11 with LMHs (1.7%). Thirty-four fellow eyes (5.2%) could not be classified and 469 fellow eyes were normal (71.3%). Furthermore, the 39 bilateral MHs were divided into four stages according to the Gass classification, as follows: stage 1, 32 eyes (82.1%); stage 2, none (0%); stage 3, 5 eyes (12.8%); and stage 4, 2 eyes (5.1%). Vitreoretinal interface abnormalities, such as MHs, ERMs, VMTS and LMHs, are more common in the fellow eyes of patients with MHs than in the healthy population. The fellow eyes of patients with MHs should be carefully followed using OCT.

Denoising OCT videos based on temporal redundancy

The identification of eye diseases and their progression often relies on a clear visualization of the anatomy and on different metrics extracted from Optical Coherence Tomography (OCT) B-scans. However, speckle noise hinders the quality of rapid OCT imaging, hampering the extraction and reliability of biomarkers that require time series. By synchronizing the acquisition of OCT images with the timing of the cardiac pulse, we transform a low-quality OCT video into a clear version by phase-wrapping each frame to the heart pulsation and averaging frames that correspond to the same instant in the cardiac cycle. Here, we compare the performance of our one-cycle denoising strategy with a deep-learning architecture, Noise2Noise, as well as classical denoising methods such as BM3D and Non-Local Means (NLM). We systematically analyze different image quality descriptors as well as region-specific metrics to assess the denoising performance based on the anatomy of the eye. The one-cycle method achieves the highest denoising performance, increases image quality and preserves the high-resolution structures within the eye tissues. The proposed workflow can be readily implemented in a clinical setting.

Macular hole with epiretinal proliferation: diagnostic value of en-face optical coherence tomography and clinical characteristics.

To elucidate the clinical features and surgical outcomes of full-thickness macular hole (FTMH) with epiretinal proliferation (EP) diagnosed by both en-face and B-mode optical coherence tomography (OCT). This retrospective cohort study classified idiopathic FTMHs into two groups, based on B-scan and en-face OCT imaging: FTMH with EP (EP group) and without EP (non-EP group). The preoperative features, as well as postoperative outcomes up to 12months, were compared between the two groups. Among 318 eyes of idiopathic FTMH that met the inclusion criteria, 59 eyes (18.6%) were in the EP group, and others were in the non-EP group. In 9 eyes (15.3%) out of the EP group, EP was not detected in the preoperative B-mode OCT but was identified through the en-face OCT. Baseline features showed a higher male proportion (47.5% vs. 27.8%, P = 0.005) and a lower incidence of vitreofoveal traction (P < 0.001) in the EP group than in the non-EP group. The EP group showed worse visual recovery than the non-EP group (- 0.23 vs. - 0.41 logarithm of the minimum angle of the resolution at 12months, P = 0.001). The en-face OCT enhances diagnostic accuracy of EP in FTMH eyes, especially in the case with smaller extent of EP. Eyes with FTMH with EP showed a worse visual recovery than FTMH without EP.

FROM OUTER RETINAL NEOVACULARIZATION TO EXUDATIVE SUBRETINAL NEOVASCULARIZATION IN MACULAR TELANGIECTASIA TYPE 2.

To describe the progression from outer retinal neovascularization (ORNV) to exudative subretinal new vessels (SRNVs) in idiopathic macular telangiectasia type 2. A total of 135 patients (270 eyes) imaged with optical coherence tomography angiography were included. Ellipsoid zone loss, outer retinal hyperreflectivity, ORNV, and SRNVs. Outer retinal neovascularization was defined as a flow signal passing through the outer plexiform layer, with or without vertical linear outer retinal hyperreflectivity on the optical coherence tomography B-scan. Subretinal new vessels were defined as an abnormal capillary network with a peripheral anastomotic arcade seen on en face optical coherence tomography angiography and a convex hyperreflectivity at the retinal pigment epithelium. Subretinal new vessels were observed in 38/270 eyes (14%). Subretinal new vessels were at a fibrotic stage in 24/38 eyes and at an exudative stage in 6/38 eyes, and a progression from ORNV to SRNVs was documented in 8/38 eyes. All cases showed an ellipsoid zone loss. In seven eyes (2.5%), SRNVs were also associated with subepithelial neovascularization. No retinochoroidal anastomosis was detected. The visual acuity dropped when SRNVs were present. In this case series, SRNVs were found in 14% of eyes. In all cases, they were associated with an ellipsoid zone loss and with outer retinal hyperreflectivity. A progression from ORNV to SRNVs was observed.

Topographic analysis of local OCT biomarkers which predict progression to atrophy in age-related macular degeneration

PurposeTo define optical coherence tomography (OCT) biomarkers that precede the development of complete retinal pigment epithelium and outer retinal atrophy (cRORA) at that location in eyes with age-related macular degeneration (AMD).MethodsIn this retrospective case–control study, patients with dry AMD who had evidence of cRORA and OCT data available for 4 years (48 ± 4 months) prior to the first visit with evidence of cRORA were included. The visit 4 years prior to the development of cRORA was defined as the baseline visit, and the region on the OCT B-scans of future cRORA development was termed the case region. A region in the same eye at the same distance from the foveal center as the case region that did not progress to cRORA was selected as the control region. OCT B-scans at the baseline visit through both the case and control regions were evaluated for the presence of soft and cuticular drusen, drusen with hyporeflective cores (hcD), drusenoid pigment epithelial detachments (PED), subretinal drusenoid deposits (SDD), thick and thin double-layer signs (DLS), intraretinal hyperreflective foci (IHRF), and acquired vitelliform lesions (AVL).ResultsA total of 57 eyes of 41 patients with dry AMD and evidence of cRORA were included. Mean time from the baseline visit to the first visit with cRORA was 44.7 ±\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\pm$$\\end{document} 6.5 months. The presence of soft drusen, drusenoid PED, AVL, thin DLS, and IHRF at the baseline visit was all associated with a significantly increased risk of cRORA at that location. Multivariable logistic regression revealed that IHRF (OR, 8.559; p < 0.001), drusenoid PED (OR, 7.148; p = 0.001), and a thin DLS (OR, 3.483; p = 0.021) were independent predictors of development of cRORA at that location.ConclusionsIHRF, drusenoid PED, and thin DLS are all local risk factors for the development of cRORA at that same location. These findings would support the inclusion of these features within a more granular staging system defining specific steps in the progression from early AMD to atrophy.

RPE Curvature Can Screen for Early and Intermediate AMD.

Diagnosing AMD early optimizes clinical management. However, current diagnostic accuracy is limited by the subjectivity of qualitative diagnostic measures used in clinical practice. This study tests if RPE curvature could be an accurate, quantitative measure for AMD diagnosis. Consecutive patients without AMD or normal aging changes (n = 111), with normal aging changes (n = 107), early AMD (n = 102) and intermediate AMD (n = 114) were recruited. RPE curvature was calculated based on the sinuosity method of measuring river curvature in environmental science. RPE and Bruch's membrane were manually segmented from optical coherence tomography B-scans and then their lengths automatically extracted using customized MATLAB code. RPE sinuosity was calculated as a ratio of RPE to Bruch's membrane length. Diagnostic accuracy was determined from area under the receiver operator characteristic curve (aROC). RPE sinuosity of foveal B-scans could distinguish any eyes with AMD (early or intermediate) from those without AMD (non-AMD or eyes with normal aging changes) with acceptable diagnostic accuracy (aROC = 0.775). Similarly, RPE sinuosity could identify intermediate AMD from all other groups (aROC = 0.871) and distinguish between early and intermediate AMD (aROC = 0.737). RPE sinuosity was significantly associated with known AMD lesions: reticular pseudodrusen (P < 0.0001) and drusen volume (P < 0.0001), but not physiological variables such as age, sex, and ethnicity. RPE sinuosity is a simple, robust, quantitative biomarker that is amenable to automation and could enhance screening of AMD.

Using clinical optical coherence tomography to characterise contact lens edge shape and base curve radius

ABSTRACT Clinical relevance Clinical optical coherence tomography devices are widely used in optometry and ophthalmology and may be used to measure contact lens base curvature radius and visualise contact lens edge shape. Background Knowledge of contact lens geometry facilitates fitting, while optical coherence tomography provides a powerful means of measuring geometrical form. This study evaluates the performance of a clinical optical coherence tomography device (3D OCT-1000) in characterising contact lens edge shape and measuring the back optic zone radius of rigid gas-permeable contact lenses in vitro. Methods First, an opto-mechanical optical coherence tomography contact lens adaptor was designed and 3D-printed to facilitate a contact lens being imaged using a commercial optical coherence tomography device. Second, several image-processing algorithms and a simple calibration method were developed to measure the back optic zone radius in optical coherence tomography B-scans. Finally, based on the findings of two experiments, B-scan performance was evaluated in terms of 1) capacity to differentiate between contact lens edge geometries, and 2) capacity to obtain accurate and repeatable back optic zone radius measurements. Statistical and graphical analyses were performed to characterise reliability and reproducibility. Results The 3D OCT-1000 and adaptor combination was capable of acquiring images of sufficient quality to discriminate between soft and rigid contact lens edge geometries. Additionally, statistical analysis of the rigid contact lens measurements demonstrated satisfactory back optic zone radius measurement accuracy and reproducibility. Conclusion This study demonstrates that a 3D OCT-1000 fitted with an opto-mechanical adaptor combination can be used to assess contact lens edges in vitro and that this clinical optical coherence tomography device, combined with image processing and linear calibration of the B-scans, is capable of obtaining back optic zone radius measurements of rigid gas-permeable contact lenses that are close to the ISO 18,369-2:2018 manufacturing tolerance range (±0.05 mm).

The characteristics of acute macular neuroretinopathy following COVID-19 infection

BackgroundIn this study, we report a case series of acute macular neuroretinopathy (AMN) associated with COVID-19 infection.MethodsThis retrospective observational study was conducted at Beijing Tongren Hospital. We reviewed patients who were diagnosed with AMN within one month of testing positive for COVID-19 using real-time reverse transcription-polymerase chain reaction (RT-PCR).ResultsA total of 11 AMN patients (20 eyes) were included in the study. The mean age was 33.8 ± 12.6 years. The average interval between a positive COVID-19 PCR test and the onset of ocular symptoms was 2.8 ± 2.5 days. The mean follow-up period for the patients was 12.5 ± 3.8 weeks. Imaging characteristics of AMN patients following COVID-19 infection included areas of low reflectivity on near-infrared reflectance (NIR) imaging, hyperreflective lesions at the level of the outer plexiform layer (OPL) and outer nuclear layer (ONL) and disruption of the ellipsoid zone (EZ) on spectral domain optical coherence tomography (SD-OCT) B-scans. Visual field examinations revealed parafoveal scotomas that closely corresponded to the clinical lesions. Optical coherence tomography angiography (OCT-A) demonstrated impaired perfusion in the deep retinal vascular plexus. Fluorescein angiography (FA), indocyanine green angiography (ICGA), and spontaneous fundus autofluorescence showed no significant abnormalities. During follow-up, partial improvement in retinal lesions was observed in NIR imaging and SD-OCT in some patients, but a proportion of patients still exhibited persistent retinal damage and no improvement in visual field scotomas.ConclusionCOVID-19-related AMN share similar clinical and imaging features with AMN due to other causes, as evidenced by the persistent presence of visual field scotomas over a longer duration.Trail registrationhttps://www.chictr.org.cn/; identifier: ChiCTR2100044365

Automated Detection of Abnormal Optical Coherence Tomography B-scans Using a Deep Learning Artificial Intelligence Neural Network Platform.

Automated Detection of Abnormal Optical Coherence Tomography B-scans Using a Deep Learning Artificial Intelligence Neural Network Platform.

A new convolutional neural network based on combination of circlets and wavelets for macular OCT classification

Artificial intelligence (AI) algorithms, encompassing machine learning and deep learning, can assist ophthalmologists in early detection of various ocular abnormalities through the analysis of retinal optical coherence tomography (OCT) images. Despite considerable progress in these algorithms, several limitations persist in medical imaging fields, where a lack of data is a common issue. Accordingly, specific image processing techniques, such as time–frequency transforms, can be employed in conjunction with AI algorithms to enhance diagnostic accuracy. This research investigates the influence of non-data-adaptive time–frequency transforms, specifically X-lets, on the classification of OCT B-scans. For this purpose, each B-scan was transformed using every considered X-let individually, and all the sub-bands were utilized as the input for a designed 2D Convolutional Neural Network (CNN) to extract optimal features, which were subsequently fed to the classifiers. Evaluating per-class accuracy shows that the use of the 2D Discrete Wavelet Transform (2D-DWT) yields superior outcomes for normal cases, whereas the circlet transform outperforms other X-lets for abnormal cases characterized by circles in their retinal structure (due to the accumulation of fluid). As a result, we propose a novel transform named CircWave by concatenating all sub-bands from the 2D-DWT and the circlet transform. The objective is to enhance the per-class accuracy of both normal and abnormal cases simultaneously. Our findings show that classification results based on the CircWave transform outperform those derived from original images or any individual transform. Furthermore, Grad-CAM class activation visualization for B-scans reconstructed from CircWave sub-bands highlights a greater emphasis on circular formations in abnormal cases and straight lines in normal cases, in contrast to the focus on irrelevant regions in original B-scans. To assess the generalizability of our method, we applied it to another dataset obtained from a different imaging system. We achieved promising accuracies of 94.5% and 90% for the first and second datasets, respectively, which are comparable with results from previous studies. The proposed CNN based on CircWave sub-bands (i.e. CircWaveNet) not only produces superior outcomes but also offers more interpretable results with a heightened focus on features crucial for ophthalmologists.