Showing papers on "Optical coherence tomography published in 2022"

Optical coherence tomography in coronary atherosclerosis assessment and intervention

Abstract: Since optical coherence tomography (OCT) was first performed in humans two decades ago, this imaging modality has been widely adopted in research on coronary atherosclerosis and adopted clinically for the optimization of percutaneous coronary intervention. In the past 10 years, substantial advances have been made in the understanding of in vivo vascular biology using OCT. Identification by OCT of culprit plaque pathology could potentially lead to a major shift in the management of patients with acute coronary syndromes. Detection by OCT of healed coronary plaque has been important in our understanding of the mechanisms involved in plaque destabilization and healing with the rapid progression of atherosclerosis. Accurate detection by OCT of sequelae from percutaneous coronary interventions that might be missed by angiography could improve clinical outcomes. In addition, OCT has become an essential diagnostic modality for myocardial infarction with non-obstructive coronary arteries. Insight into neoatherosclerosis from OCT could improve our understanding of the mechanisms of very late stent thrombosis. The appropriate use of OCT depends on accurate interpretation and understanding of the clinical significance of OCT findings. In this Review, we summarize the state of the art in cardiac OCT and facilitate the uniform use of this modality in coronary atherosclerosis. Contributions have been made by clinicians and investigators worldwide with extensive experience in OCT, with the aim that this document will serve as a standard reference for future research and clinical application. Optical coherence tomography (OCT) has been widely adopted in research on coronary atherosclerosis and adopted clinically to optimize percutaneous coronary intervention. In this Review, Jang and colleagues summarize this rapidly progressing field, with the aim of standardizing the use of OCT in coronary atherosclerosis.

An overview of artificial intelligence in diabetic retinopathy and other ocular diseases

Abstract: Artificial intelligence (AI), also known as machine intelligence, is a branch of science that empowers machines using human intelligence. AI refers to the technology of rendering human intelligence through computer programs. From healthcare to the precise prevention, diagnosis, and management of diseases, AI is progressing rapidly in various interdisciplinary fields, including ophthalmology. Ophthalmology is at the forefront of AI in medicine because the diagnosis of ocular diseases heavy reliance on imaging. Recently, deep learning-based AI screening and prediction models have been applied to the most common visual impairment and blindness diseases, including glaucoma, cataract, age-related macular degeneration (ARMD), and diabetic retinopathy (DR). The success of AI in medicine is primarily attributed to the development of deep learning algorithms, which are computational models composed of multiple layers of simulated neurons. These models can learn the representations of data at multiple levels of abstraction. The Inception-v3 algorithm and transfer learning concept have been applied in DR and ARMD to reuse fundus image features learned from natural images (non-medical images) to train an AI system with a fraction of the commonly used training data (<1%). The trained AI system achieved performance comparable to that of human experts in classifying ARMD and diabetic macular edema on optical coherence tomography images. In this study, we highlight the fundamental concepts of AI and its application in these four major ocular diseases and further discuss the current challenges, as well as the prospects in ophthalmology.

A Novel Computer-Aided Diagnostic System for Early Detection of Diabetic Retinopathy Using 3D-OCT Higher-Order Spatial Appearance Model

Abstract: Early diagnosis of diabetic retinopathy (DR) is of critical importance to suppress severe damage to the retina and/or vision loss. In this study, an optical coherence tomography (OCT)-based computer-aided diagnosis (CAD) method is proposed to detect DR early using structural 3D retinal scans. This system uses prior shape knowledge to automatically segment all retinal layers of the 3D-OCT scans using an adaptive, appearance-based method. After the segmentation step, novel texture features are extracted from the segmented layers of the OCT B-scans volume for DR diagnosis. For every layer, Markov–Gibbs random field (MGRF) model is used to extract the 2nd-order reflectivity. In order to represent the extracted image-derived features, we employ cumulative distribution function (CDF) descriptors. For layer-wise classification in 3D volume, using the extracted Gibbs energy feature, an artificial neural network (ANN) is fed the extracted feature for every layer. Finally, the classification outputs for all twelve layers are fused using a majority voting schema for global subject diagnosis. A cohort of 188 3D-OCT subjects are used for system evaluation using different k-fold validation techniques and different validation metrics. Accuracy of 90.56%, 93.11%, and 96.88% are achieved using 4-, 5-, and 10-fold cross-validation, respectively. Additional comparison with deep learning networks, which represent the state-of-the-art, documented the promise of our system’s ability to diagnose the DR early.

Diagnosis of Retinal Diseases Based on Bayesian Optimization Deep Learning Network Using Optical Coherence Tomography Images

Abstract: Retinal abnormalities have emerged as a serious public health concern in recent years and can manifest gradually and without warning. These diseases can affect any part of the retina, causing vision impairment and indeed blindness in extreme cases. This necessitates the development of automated approaches to detect retinal diseases more precisely and, preferably, earlier. In this paper, we examine transfer learning of pretrained convolutional neural network (CNN) and then transfer it to detect retinal problems from Optical Coherence Tomography (OCT) images. In this study, pretrained CNN models, namely, VGG16, DenseNet201, InceptionV3, and Xception, are used to classify seven different retinal diseases from a dataset of images with and without retinal diseases. In addition, to choose optimum values for hyperparameters, Bayesian optimization is applied, and image augmentation is used to increase the generalization capabilities of the developed models. This research also provides a comparison of the proposed models as well as an analysis of them. The accuracy achieved using DenseNet201 on the Retinal OCT Image dataset is more than 99% and offers a good level of accuracy in classifying retinal diseases compared to other approaches, which only detect a small number of retinal diseases.

Line-Field Confocal Optical Coherence Tomography: A New Tool for the Differentiation between Nevi and Melanomas?

Abstract: Simple Summary Typical benign nevi and advanced melanomas can be easily discriminated, but there are still some melanocytic lesions where even experts are not sure about the correct diagnosis and degree of malignity. The high penetration depth of optical coherence tomography (OCT) allows an assessment of tumor thickness of the lesion precisely, but without cellular resolution the differentiation of melanocytic lesions remains difficult. On the other hand, reflectance confocal microscopy (RCM) allows for very good morphological identification of either a nevus or a melanoma, but cannot show the infiltration depth of the lesion because of its low penetration depth. Since the new device of line-field confocal optical coherence tomography (LC-OCT) technically closes the gap between these other two devices, in this study, we wanted to examine if it is possible to differentiate between nevi and melanomas with LC-OCT, and which criteria are the most important for it. Abstract Until now, the clinical differentiation between a nevus and a melanoma is still challenging in some cases. Line-field confocal optical coherence tomography (LC-OCT) is a new tool with the aim to change that. The aim of the study was to evaluate LC-OCT for the discrimination between nevi and melanomas. A total of 84 melanocytic lesions were examined with LC-OCT and 36 were also imaged with RCM. The observers recorded the diagnoses, and the presence or absence of the 18 most common imaging parameters for melanocytic lesions, nevi, and melanomas in the LC-OCT images. Their confidence in diagnosis and the image quality of LC-OCT and RCM were evaluated. The most useful criteria, the sensitivity and specificity of LC-OCT vs. RCM vs. histology, to differentiate a (dysplastic) nevus from a melanoma were analyzed. Good image quality correlated with better diagnostic performance (Spearman correlation: 0.4). LC-OCT had a 93% sensitivity and 100% specificity compared to RCM (93% sensitivity, 95% specificity) for diagnosing a melanoma (vs. all types of nevi). No difference in performance between RCM and LC-OCT was observed (McNemar’s p value = 1). Both devices falsely diagnosed dysplastic nevi as non-dysplastic (43% sensitivity for dysplastic nevus diagnosis). The most significant criteria for diagnosing a melanoma with LC-OCT were irregular honeycombed patterns (92% occurrence rate; 31.7 odds ratio (OR)), the presence of pagetoid spread (89% occurrence rate; 23.6 OR) and the absence of dermal nests (23% occurrence rate, 0.02 OR). In conclusion LC-OCT is useful for the discrimination between melanomas and nevi.

Metasurface-based bijective illumination collection imaging provides high-resolution tomography in three dimensions

Abstract: Microscopic imaging in three dimensions enables numerous biological and clinical applications. However, high-resolution optical imaging preserved in a relatively large depth range is hampered by the rapid spread of tightly confined light due to diffraction. Here, we show that a particular disposition of light illumination and collection paths liberates optical imaging from the restrictions imposed by diffraction. This arrangement, realized by metasurfaces, decouples lateral resolution from depth-of-focus by establishing a one-to-one correspondence (bijection) along a focal line between the incident and collected light. Implementing this approach in optical coherence tomography, we demonstrate tissue imaging at 1.3 μm wavelength with ~ 3.2 μm lateral resolution, maintained nearly intact over 1.25 mm depth-of-focus, with no additional acquisition or computation burden. This method, termed bijective illumination collection imaging, is general and might be adapted across various existing imaging modalities.

In Vivo Imaging of Retinal and Choroidal Morphology and Vascular Plexuses of Vertebrates Using Swept-Source Optical Coherence Tomography

Abstract: Purpose To perform in vivo evaluation of the structural morphology and vascular plexuses of the neurosensory retina and choroid across vertebrate species using swept-source optical coherence tomography (SS-OCT) and SS-OCT angiography (SS-OCTA) imaging. Methods A custom-built SS-OCT system with an incorporated flexible imaging arm was used to acquire the three-dimensional (3D) retinal OCT and vascular OCTA data of five different vertebrates: a mouse (C57BL/6J), a rat (Long Evans), a gray short-tailed opossum (Monodelphis domestica), a white sturgeon (Acipenser transmontanus), and a great horned owl (Bubo virginianus). Results In vivo structural morphology of the retina and choroid, as well as en face OCTA images of retinal and choroidal vasculature of all species were generated. The retinal morphology and vascular plexuses were similar between rat and mouse, whereas distinct choroidal and paired superficial vessels were observed in the opossum retina. The retinal and vascular structure of the sturgeon, as well as the pecten oculi and overlying the avascular and choroidal vasculature in the owl retina are reported in vivo. Conclusions A high-quality two-dimensional and 3D in vivo visualization of the retinal structures and en face visualization of the retina and choroidal vascular plexus of vertebrates was possible. Our studies affirm that SS-OCT and SS-OCTA are viable methods for evaluating the in vivo retinal and choroidal structure across terrestrial, aquatic, and aerial vertebrates. Translational Relevance In vivo characterization of retinal morphology and vasculature plexus of multiple species using SS-OCT and SS-OCTA imaging can increase the pool of species available as models of human retinal diseases.

Deep Learning Prediction of Response to Anti-VEGF among Diabetic Macular Edema Patients: Treatment Response Analyzer System (TRAS)

Abstract: Diabetic macular edema (DME) is the most common cause of visual impairment among patients with diabetes mellitus. Anti-vascular endothelial growth factors (Anti-VEGFs) are considered the first line in its management. The aim of this research has been to develop a deep learning (DL) model for predicting response to intravitreal anti-VEGF injections among DME patients. The research included treatment naive DME patients who were treated with anti-VEGF. Patient’s pre-treatment and post-treatment clinical and macular optical coherence tomography (OCT) were assessed by retina specialists, who annotated pre-treatment images for five prognostic features. Patients were also classified based on their response to treatment in their post-treatment OCT into either good responder, defined as a reduction of thickness by >25% or 50 µm by 3 months, or poor responder. A novel modified U-net DL model for image segmentation, and another DL EfficientNet-B3 model for response classification were developed and implemented for predicting response to anti-VEGF injections among patients with DME. Finally, the classification DL model was compared with different levels of ophthalmology residents and specialists regarding response classification accuracy. The segmentation deep learning model resulted in segmentation accuracy of 95.9%, with a specificity of 98.9%, and a sensitivity of 87.9%. The classification accuracy of classifying patients’ images into good and poor responders reached 75%. Upon comparing the model’s performance with practicing ophthalmology residents, ophthalmologists and retina specialists, the model’s accuracy is comparable to ophthalmologist’s accuracy. The developed DL models can segment and predict response to anti-VEGF treatment among DME patients with comparable accuracy to general ophthalmologists. Further training on a larger dataset is nonetheless needed to yield more accurate response predictions.

Speckle Noise Reduction for OCT Images Based on Image Style Transfer and Conditional GAN

Abstract: Raw optical coherence tomography (OCT) images typically are of low quality because speckle noise blurs retinal structures, severely compromising visual quality and degrading performances of subsequent image analysis tasks. In our previous study (Ma et al., 2018), we have developed a Conditional Generative Adversarial Network (cGAN) for speckle noise removal in OCT images collected by several commercial OCT scanners, which we collectively refer to as scanner T. In this paper, we improve the cGAN model and apply it to our in-house OCT scanner (scanner B) for speckle noise suppression. The proposed model consists of two steps: 1) We train a Cycle-Consistent GAN (CycleGAN) to learn style transfer between two OCT image datasets collected by different scanners. The purpose of the CycleGAN is to leverage the ground truth dataset created in our previous study. 2) We train a mini-cGAN model based on the PatchGAN mechanism with the ground truth dataset to suppress speckle noise in OCT images. After training, we first apply the CycleGAN model to convert raw images collected by scanner B to match the style of the images from scanner T, and subsequently use the mini-cGAN model to suppress speckle noise in the style transferred images. We evaluate the proposed method on a dataset collected by scanner B. Experimental results show that the improved model outperforms our previous method and other state-of-the-art models in speckle noise removal, retinal structure preservation and contrast enhancement.

Line-Field Confocal Optical Coherence Tomography Increases the Diagnostic Accuracy and Confidence for Basal Cell Carcinoma in Equivocal Lesions: A Prospective Study

Abstract: Simple Summary Basal cell carcinoma is the most frequently occurring type of skin cancer. Its treatment can be either local or surgical depending on its subtype and extension, with early recognized and superficial cases being easier to treat. Some of them, however, display unspecific features, making diagnosis difficult. Non-invasive devices such as line-field confocal optical coherence tomography (LC-OCT) are able to recognize morphological features of different BCC subtypes with a good correlation to histopathology. We decided to study their application to clinically doubtful BCC cases. Abstract Diagnosing clinically unclear basal cell carcinomas (BCCs) can be challenging. Line-field confocal optical coherence tomography (LC-OCT) is able to display morphological features of BCC subtypes with good histological correlation. The aim of this study was to investigate the accuracy of LC-OCT in diagnosing clinically unsure cases of BCC compared to dermoscopy alone and in distinguishing between superficial BCCs and other BCC subtypes. Moreover, we addressed pitfalls in false positive cases. We prospectively enrolled 182 lesions of 154 patients, referred to our department to confirm or to rule out the diagnosis of BCC. Dermoscopy and LC-OCT images were evaluated by two experts independently. Image quality, LC-OCT patterns and criteria, diagnosis, BCC subtype, and diagnostic confidence were assessed. Sensitivity and specificity of additional LC-OCT were compared to dermoscopy alone for identifying BCC in clinically unclear lesions. In addition, key LC-OCT features to distinguish between BCCs and non-BCCs and to differentiate superficial BCCs from other BCC subtypes were determined by linear regressions. Diagnostic confidence was rated as “high” in only 48% of the lesions with dermoscopy alone compared to 70% with LC-OCT. LC-OCT showed a high sensitivity (98%) and specificity (80%) compared to histology, and these were even higher (100% sensitivity and 97% specificity) in the subgroup of lesions with high diagnostic confidence. Interobserver agreement was nearly perfect (95%). The combination of dermoscopy and LC-OCT reached a sensitivity of 100% and specificity of 81.2% in all cases and increased to sensitivity of 100% and specificity of 94.9% in cases with a high diagnostic confidence. The performance of LC-OCT was influenced by the image quality but not by the anatomical location of the lesion. The most specific morphological LC-OCT criteria in BCCs compared to non-BCCs were: less defined dermoepidermal junction (DEJ), hyporeflective tumor lobules, and dark rim. The most relevant features of the subgroup of superficial BCCs (sBCCs) were: string of pearls pattern and absence of epidermal thinning. Our diagnostic confidence, sensitivity, and specificity in detecting BCCs in the context of clinically equivocal lesions significantly improved using LC-OCT in comparison to dermoscopy only. Operator training for image acquisition is fundamental to achieve the best results. Not only the differential diagnosis of BCC, but also BCC subtyping can be performed at bedside with LC-OCT.

Inflation of test accuracy due to data leakage in deep learning-based classification of OCT images

Abstract: In the application of deep learning on optical coherence tomography (OCT) data, it is common to train classification networks using 2D images originating from volumetric data. Given the micrometer resolution of OCT systems, consecutive images are often very similar in both visible structures and noise. Thus, an inappropriate data split can result in overlap between the training and testing sets, with a large portion of the literature overlooking this aspect. In this study, the effect of improper dataset splitting on model evaluation is demonstrated for three classification tasks using three OCT open-access datasets extensively used, Kermany's and Srinivasan's ophthalmology datasets, and AIIMS breast tissue dataset. Results show that the classification performance is inflated by 0.07 up to 0.43 in terms of Matthews Correlation Coefficient (accuracy: 5% to 30%) for models tested on datasets with improper splitting, highlighting the considerable effect of dataset handling on model evaluation. This study intends to raise awareness on the importance of dataset splitting given the increased research interest in implementing deep learning on OCT data.

Choroidal Thickness and Its Association With Age, Axial Length, and Refractive Error in Chinese Adults

Abstract: Purpose To identify the association between the choroidal thickness (ChT) with age and axial length (AL) under different refractive errors (REs) in Chinese adults. Methods Swept-source optical coherence tomography was used to measure ChT in 2126 right eyes of 2126 participants. The participants were classified as having pathologic myopia (PM), high myopia without PM (HM), low myopia (LM), and nonmyopia (non-M) according to their REs and META-PM (the Meta-Analysis of Pathologic Myopia) classification criteria. Results The mean age was 52.49 ± 20.39 years (range, 18−93 years), and the mean RE was −5.27 ± 5.37 diopters (D; range, −25.5 to +7.75 D). The mean average ChT was 159.25 ± 80.75 µm and decreased in a linear relationship from non-M to PM (190.04 ± 72.64 µm to 60.99 ± 37.58 µm, P < 0.001). A significant decline in ChT was noted between 50 and 70 years (r = −0.302, P < 0.001) and less rapidly after the age of 70 years (r = −0.105, P = 0.024). No correlation was noted between age and ChT under 50 years (P = 0.260). A significantly higher association with AL was noted in the central fovea (βHM = −23.92, βLM = −23.88, βNon-M = −18.80, all P < 0.001) and parafoveal ChT (βHM = −22.87, βLM = −22.31, βNon-M = −18.61, all P < 0.001) when compared with the perifoveal region (βHM = −19.80, βLM = −18.29, βNon-M = −13.95, all P < 0.001). Within each group of PM, HM, LM, and non-M, regression analysis showed that the coefficients of age and AL with different macular regions of ChT varied significantly. Conclusions ChT was negatively correlated with age after 50 years. The thinning of the choroid was more prominent in the center and parafoveal regions as AL increased. Varied distributions of ChT decrease associated with AL and age were noted among different refractive groups.

3D-Printed Micro Lens-in-Lens for In Vivo Multimodal Microendoscopy.

Abstract: Multimodal microendoscopes enable co-located structural and molecular measurements in vivo, thus providing useful insights into the pathological changes associated with disease. However, different optical imaging modalities often have conflicting optical requirements for optimal lens design. For example, a high numerical aperture (NA) lens is needed to realize high-sensitivity fluorescence measurements. In contrast, optical coherence tomography (OCT) demands a low NA to achieve a large depth of focus. These competing requirements present a significant challenge in the design and fabrication of miniaturized imaging probes that are capable of supporting high-quality multiple modalities simultaneously. An optical design is demonstrated which uses two-photon 3D printing to create a miniaturized lens that is simultaneously optimized for these conflicting imaging modalities. The lens-in-lens design contains distinct but connected optical surfaces that separately address the needs of both fluorescence and OCT imaging within a lens of 330 µm diameter. This design shows an improvement in fluorescence sensitivity of >10x in contrast to more conventional fiber-optic design approaches. This lens-in-lens is then integrated into an intravascular catheter probe with a diameter of 520 µm. The first simultaneous intravascular OCT and fluorescence imaging of a mouse artery in vivo is reported.

Evaluation of a self-imaging SD-OCT system designed for remote home monitoring

Abstract: Abstract Purpose To compare identification rates of retinal fluid of the Notal Vision Home Optical Coherence Tomography (OCT) device (NVHO) when used by people with age-related macular degeneration (AMD) to those captured by a commercial OCT. Methods Prospective, cross-sectional study where patients underwent commercial OCT imaging followed by self-imaging with either the NVHO 2.5 or the NVHO 3 in clinic setting. Outcomes included patients’ ability to acquire analyzable OCT images with the NVHO and to compare those with commercial images. Results Successful images were acquired with the NVHO 2.5 in 469/531 eyes (88%) in 264/290 subjects (91%) with the mean (SD) age of 78.8 (8.8); 153 (58%) were female with median visual acuity (VA) of 20/40. In the NVHO 3 cohort, 69 eyes of 45 subjects (93%) completed the self-imaging. Higher rates of successful imaging were found in eyes with VA ≥ 20/320. Positive percent agreement/negative percent agreement for detecting the presence of subretinal and/or intraretinal fluid when reviewing for fluid in three repeated volume scans were 97%/95%, respectively for the NVHO v3. Conclusion Self-testing with the NVHO can produce high quality images suitable for fluid identification by human graders, suggesting the device may be able to complement standard-of-care clinical assessments and treatments.

Non-invasive scoring of cellular atypia in keratinocyte cancers in 3D LC-OCT images using Deep Learning

Abstract: Diagnosis based on histopathology for skin cancer detection is today's gold standard and relies on the presence or absence of biomarkers and cellular atypia. However it suffers drawbacks: it requires a strong expertise and is time-consuming. Moreover the notion of atypia or dysplasia of the visible cells used for diagnosis is very subjective, with poor inter-rater agreement reported in the literature. Lastly, histology requires a biopsy which is an invasive procedure and only captures a small sample of the lesion, which is insufficient in the context of large fields of cancerization. Here we demonstrate that the notion of cellular atypia can be objectively defined and quantified with a non-invasive in-vivo approach in three dimensions (3D). A Deep Learning (DL) algorithm is trained to segment keratinocyte (KC) nuclei from Line-field Confocal Optical Coherence Tomography (LC-OCT) 3D images. Based on these segmentations, a series of quantitative, reproducible and biologically relevant metrics is derived to describe KC nuclei individually. We show that, using those metrics, simple and more complex definitions of atypia can be derived to discriminate between healthy and pathological skins, achieving Area Under the ROC Curve (AUC) scores superior than 0.965, largely outperforming medical experts on the same task with an AUC of 0.766. All together, our approach and findings open the door to a precise quantitative monitoring of skin lesions and treatments, offering a promising non-invasive tool for clinical studies to demonstrate the effects of a treatment and for clinicians to assess the severity of a lesion and follow the evolution of pre-cancerous lesions over time.

Morphological evaluation of melanocytic lesions with three‐dimensional line‐field confocal optical coherence tomography: correlation with histopathology and reflectance confocal microscopy. A pilot study

Abstract: Line‐field confocal optical coherence tomography (LC‐OCT) is a new in vivo emerging technique that provides cellular resolution, allows deep imaging (400 μm) and produces real‐time images in both the horizontal and vertical plane and in three dimensions. No previous description of different subtypes of melanocytic lesions and their correlation with histopathology and reflectance confocal microscopy has been reported.

Line‐field confocal optical coherence tomography of psoriasis, eczema and lichen planus: a case series with histopathological correlation

Abstract: Line‐field confocal optical coherence tomography (LC‐OCT) is a novel, non‐invasive technique that provides in vivo, high‐resolution images in both vertical and horizontal sections.

Two Cases of Acute Macular Neuroretinopathy Associated with the Adenovirus-based COVID-19 Vaccine Vaxzevria (Astrazeneca)

Abstract: ABSTRACT Purpose To report two cases of acute macular neuroretinopathy (AMN) in young female patients following the administration of the adenovirus-based coronavirus disease 2019 (COVID-19) vaccine Vaxzevria (AstraZeneca). Methods Spectral-domain optical coherence tomography and infrared imaging were used to confirm the diagnosis of AMN. Results Both patients showed a parafoveal hyperreflective band in the outer nuclear layer, disruption of the ellipsoid and interdigitation zones of the photoreceptor layers, and correlating hyporeflective areas on the near-infrared images. Both patients presented with flu-like fever and sudden onset of fortifications within 48 hours of vaccination. One patient showed altered flow in the deep capillary plexus and highly elevated thrombotic parameters. Conclusion We report a possible association between immune-mediated AMN and the administration of adenovirus-based COVID-19 vaccine Vaxzevria.

Automatic geographic atrophy segmentation using optical attenuation in OCT scans with deep learning.

Abstract: A deep learning algorithm was developed to automatically identify, segment, and quantify geographic atrophy (GA) based on optical attenuation coefficients (OACs) calculated from optical coherence tomography (OCT) datasets. Normal eyes and eyes with GA secondary to age-related macular degeneration were imaged with swept-source OCT using 6 × 6 mm scanning patterns. OACs calculated from OCT scans were used to generate customized composite en face OAC images. GA lesions were identified and measured using customized en face sub-retinal pigment epithelium (subRPE) OCT images. Two deep learning models with the same U-Net architecture were trained using OAC images and subRPE OCT images. Model performance was evaluated using DICE similarity coefficients (DSCs). The GA areas were calculated and compared with manual segmentations using Pearson's correlation and Bland-Altman plots. In total, 80 GA eyes and 60 normal eyes were included in this study, out of which, 16 GA eyes and 12 normal eyes were used to test the models. Both models identified GA with 100% sensitivity and specificity on the subject level. With the GA eyes, the model trained with OAC images achieved significantly higher DSCs, stronger correlation to manual results and smaller mean bias than the model trained with subRPE OCT images (0.940 ± 0.032 vs 0.889 ± 0.056, p = 0.03, paired t-test, r = 0.995 vs r = 0.959, mean bias = 0.011 mm vs mean bias = 0.117 mm). In summary, the proposed deep learning model using composite OAC images effectively and accurately identified, segmented, and quantified GA using OCT scans.

Optic Nerve Drusen Evaluation: A Comparison between Ultrasound and OCT

Abstract: This observational study compared optic coherence tomography (OCT) and B-scan in the detection of optic disc drusen. In total, 86 eyes of 50 patients with optic disc drusen (ODD) (36 bilateral) with a mean age of 34.68 ± 23.81 years, and 54 eyes of 27 patients with papilledema, with a mean age of 35.42 years ± 17.47, were examined. Patients with ODD, diagnosed with ultrasound, underwent spectral-domain OCT evaluation. With US, 28 ODD cases were classified as large (4 buried and 24 superficial), 58 were classified as point-like (6 buried, 49 superficial and 3 mixed). Then, all patients underwent OCT. OCT was able to detect the presence of ODD and/or peripapillary hyperreflective ovoid mass structure (PHOMS) in 69 eyes (p < 0.001). In particular, 7 eyes (8.14%) showed the presence of ODD alone, 25 eyes (29.07%) showed only PHOMS and 37 eyes (43.02%) showed ODD and PHOMS. In 17 eyes (19.77%) no ODD or PHOMS were detected. In the papilledema group, no ODD were observed with both US and OCT. OCT showed the presence of drusen or similar lesions in only 80.23% of the cases highlighted by the US scan, so it does not allow for certain ODD diagnoses, especially in the case of buried ODD.

Comparison of Machine Learning Methods Using Spectralis OCT for Diagnosis and Disability Progression Prognosis in Multiple Sclerosis

Abstract: Machine learning approaches in diagnosis and prognosis of multiple sclerosis (MS) were analysed using retinal nerve fiber layer (RNFL) thickness, measured by optical coherence tomography (OCT). A cross-sectional study (72 MS patients and 30 healthy controls) was used for diagnosis. These 72 MS patients were involved in a 10-year longitudinal follow-up study for prognostic purposes. Structural measurements of RNFL thickness were performed using different Spectralis OCT protocols: fast macular thickness protocol to measure macular RNFL, and fast RNFL thickness protocol and fast RNFL-N thickness protocol to measure peripapillary RNFL. Binary classifiers such as multiple linear regression (MLR), support vector machines (SVM), decision tree (DT), k-nearest neighbours (k-NN), Naïve Bayes (NB), ensemble classifier (EC) and long short-term memory (LSTM) recurrent neural network were tested. For MS diagnosis, the best acquisition protocol was fast macular thickness protocol using k-NN (accuracy: 95.8%; sensitivity: 94.4%; specificity: 97.2%; precision: 97.1%; AUC: 0.958). For MS prognosis, our model with a 3-year follow up to predict disability progression 8 years later was the best predictive model. DT performed best for fast macular thickness protocol (accuracy: 91.3%; sensitivity: 90.0%; specificity: 92.5%; precision: 92.3%; AUC: 0.913) and SVM for fast RNFL-N thickness protocol (accuracy: 91.3%; sensitivity: 87.5%; specificity: 95.0%; precision: 94.6%; AUC: 0.913). This work concludes that measurements of RNFL thickness obtained with Spectralis OCT have a good ability to diagnose MS and to predict disability progression in MS patients. This machine learning approach would help clinicians to have valuable information.

In Vivo Sublayer Analysis of Human Retinal Inner Plexiform Layer Obtained by Visible-Light Optical Coherence Tomography

Abstract: Growing evidence suggests that dendrite retraction or degeneration in a subpopulation of the retinal ganglion cells (RGCs) may precede detectable soma abnormalities and RGC death in glaucoma. Visualization of the lamellar structure of the inner plexiform layer (IPL) could advance clinical management and fundamental understanding of glaucoma. We investigated whether visible-light optical coherence tomography (vis-OCT) could detect the difference in the IPL sublayer thicknesses between small cohorts of healthy and glaucomatous subjects.We imaged nine healthy and five glaucomatous subjects with vis-OCT. Four of the healthy subjects were scanned three times each in two separate visits, and five healthy and five glaucoma subjects were scanned three times during a single visit. IPL sublayers were manually segmented using averaged A-line profiles.The mean ages of glaucoma and healthy subjects are 59.6 ± 13.4 and 45.4 ± 14.4 years (P = 0.02.) The visual field mean deviations (MDs) are -26.4 to -7.7 dB in glaucoma patients and -1.6 to 1.1 dB in healthy subjects (P = 0.002). Median coefficients of variation (CVs) of intrasession repeatability for the entire IPL and three sublayers are 3.1%, 5.6%, 6.9%, and 5.6% in healthy subjects and 1.8%, 6.0%, 7.7%, and 6.2% in glaucoma patients, respectively. The mean IPL thicknesses are 36.2 ± 1.5 µm in glaucomatous and 40.1 ± 1.7 µm in healthy eyes (P = 0.003).IPL sublayer analysis revealed that the middle sublayer could be responsible for the majority of IPL thinning in glaucoma. Vis-OCT quantified IPL sublayers with good repeatability in both glaucoma and healthy subjects.