A robust deep learning classifier for screening multiple retinal diseases on optical coherence tomography

Structural Features Associated With the Development and Progression of RORA Secondary to Maternally Inherited Diabetes and Deafness

OCT Interpretations

In this paper, five types of tissues, human enamel, human cortical bone, human trabecular bone, muscular tissue, and fatty tissue were imaged <i>ex vivo</i> using optical coherence tomography (OCT). The specimens were prepared in blocks of 5 x 5 x 3 mm (width x length x height). The OCT imaging system was a swept source OCT system operating at wavelengths ranging between 1250 nm and 1360 nm with an average power of 18 mW and a scan rate of 50 to 100 kHz. The imaging probe was placed on top of a 2 x 2 cm stabilizing device to maintain a standard distance from the samples. Ten image samples from each type of tissue were obtained. To acquire images with minimum inhomogeneity, imaging was performed multiple times at different points. Based on the observed texture differences between OCT images of soft and hard tissues, spatial and spectral features were quantitatively extracted from the OCT images. The Radon transform from angles of 0 deg to 90 deg was computed, averaged over all the angles, normalized to peak at unity, and then fitted with Gaussian function. The mean absolute values of the spatial frequency components of the OCT image were considered as a feature, where 2-D fast Fourier transform (FFT) was done to OCT images. These OCT features can reliably differentiate between a range of hard and soft tissues, and could be extremely valuable in assisting dentists for <i>in vivo</i> evaluation of oral tissues and early detection of pathologic changes in tissues.

OCT Patterns in Ulcerative Colitis

To evaluate the optical coherence tomographic (OCT) features of the retina of rats, we compared the OCT images with the histological appearance of normal retinas and retinas with laser-induced choroidal neovascularization. Twelve eyes of 12 adult pigmented rats (Brown-Norway) were used. Color fundus photography, fluorescein angiography (FAG), and OCT images of normal retinas and retinas with laser photocoagulation-induced choroidal neovascularization were studied. OCT showed a double-layered structure in the normal sensory retina with a highly reflective layer located in the inner retina and a low reflective layer located in the outer retina. The retinal pigment epithelium (RPE) and choriocapillaris were imaged as a layer with the highest reflection. On the first day after photocoagulation, OCT showed a disruption of the highly reflective layer corresponding to the RPE, and an enhanced reflectivity in the choroid under the lesion. Choroidal neovascularizations (CNVs) which appeared 2 weeks after photocoagulation was seen as a multi-layered, highly reflective area protruding from the RPE into the subretinal space A CNV beneath a subretinal hematoma was difficult to detect because of the low transmission of the scanning light through the hematoma. The histopathological appearance was well correlated with the OCT images. The two reflective bands in the OCT images were identified as coming from the inner layers of the retina and from the photoreceptors. The highest reflective band arose from the RPE and choriocapillaris. In the future, OCT combined with FAG or indocyanine-green angiography will be a useful tool for the evaluation of animal studies of choroidal neovascularization and other retinal diseases.

Retinal diseases become more complex for people at any age. In the early stages, many people suffering from retinal diseases have very mild symptoms. It is observed that retinal diseases mainly damage the blood vessels that cause the leakage of fluid. The accretion of fluid can impact the retina and cause vision changes. Retinal diseases such as diabetic retinopathy (DR), age-related macular degeneration (AMD), diabetic macular edema (DME), drusen, and choroidal neovascularization (CNV) are complex diseases that show a huge impact on human retinal health. If the patients are not detected with retinal diseases in the early stages, this may lead to permanent vision loss. These diseases have a lot of other side effects, such as brain disorders. Prevention of these diseases may stop permanent vision loss in patients. Machine learning (ML) algorithms are most widely used to detect retinal diseases in their early stages. The main disadvantage of ML is that these algorithms will take more time to process the data and select complex methods. In this paper, deep learning (DL) algorithms are discussed, and various optical coherence tomography (OCT) datasets are used for experiments. Experimental results show the performance of various ML and DL approaches applied to OCT and retinal image datasets. The performance and comparison of various algorithms is also discussed in this paper.

Area under the curve may hide poor generalisation to external datasets

Objective To investigate the imaging features of optical coherence tomography (OCT) in diabetic retinopathy (DR). Methods From January 2015 to December 2017, 104 cases (208 eyes) with DR in the Ophthalmology Department of Inner Mongolia Hospital of Traditional Chinese Medicine were collected, and 289 cases (578 eyes) with DR in Tongren Ophthalmology Center of Beijing Tongren Hospital affiliated to Capital Medical University were collected. A total of 393 cases (786 eyes) were included in the two sites. Among them, there were 181 cases (356 eyes) and 212 cases (430 eyes). The age was 22~88 years, with an average age of (47.30 + 5.42) years. Name, sex, age, address, diabetes history, clinical diagnosis, fundus color photography and OCT report were recorded. Summarize and extract features in each phase of OCT inspection. Results According to the clinical staging standard of DR in 2002, 393 cases (786 eyes) were divided into three groups: 51 cases (102 eyes), 83 cases (166 eyes), 124 cases (248 eyes), 46 cases (92 eyes), 57 cases (114 eyes) and 32 cases (64 eyes) of DR stage VI. Eye only. The OCT of DR stage I was characterized by microangioma, occasional or minor bleeding spots between the retinal neuroepithelial layers; the OCT of DR stage II was characterized by rigid exudation and bleeding spots between the retinal neuroepithelial layers; the OCT of DR stage III was characterized by cotton flocculent plaques in addition to hard exudation and bleeding spots; and the OCT of DR stage IV was characterized by cotton flocculent plaques. In addition to hard exudation, bleeding spots and cotton wadding spots, there are also edema or cystoid edema and/or vitreous hemorrhage; in DRV stage, the OCT features are not only hard exudation, bleeding spots, cotton wadding spots, edema and vitreous hemorrhage, but also organic film; in DRVI stage, the OCT features are hard exudation, bleeding spots and vitreous hemorrhage. There were also traction retinal detachment outside cotton spots, edema, vitreous hemorrhage and organic membrane. In addition, according to the different stages of DR, blood flow OCT imaging (OCTA) can clearly show the characteristics of microangioma, hard exudation, hemorrhage, cystic edema, no perfusion area and neovascularization. Conclusion In the examination results of OCT, DR can appear characteristic image according to different stages. Mastering and using OCT can determine the clinical stages of DR and guide the clinical diagnosis and treatment. Key words: Diabetic retinopathy; Clinical stages; Diagnose; Optical coherence tomography

Microaneurysms (MAs) are pathognomonic signs that help clinicians to detect diabetic retinopathy (DR) in the early stages. Automatic detection of MA in retinal images is an active area of research due to its application in screening processes for DR which is one of the main reasons of blindness amongst the working-age population. The focus of these works is on the automatic detection of MAs in en face retinal images like fundus color and Fluorescein Angiography (FA). On the other hand, detection of MAs from Optical Coherence Tomography (OCT) images has 2 main advantages: first, OCT is a non-invasive imaging technique that does not require injection, therefore is safer. Secondly, because of the proven application of OCT in detection of Age-Related Macular Degeneration, Diabetic Macular Edema, and normal cases, thanks to detecting MAs in OCT, extensive information is obtained by using this imaging technique. In this research, the concentration is on the diagnosis of MAs using deep learning in the OCT images which represent in-depth structure of retinal layers. To this end, OCT B-scans should be divided into strips and MA patterns should be searched in the resulted strips. Since we need a dataset comprising OCT image strips with suitable labels and such large labelled datasets are not yet available, we have created it. For this purpose, an exact registration method is utilized to align OCT images with FA photographs. Then, with the help of corresponding FA images, OCT image strips are created from OCT B-scans in four labels, namely MA, normal, abnormal, and vessel. Once the dataset of image strips is prepared, a stacked generalization (stacking) ensemble of four fine-tuned, pre-trained convolutional neural networks is trained to classify the strips of OCT images into the mentioned classes. FA images are used once to create OCT strips for training process and they are no longer needed for subsequent steps. Once the stacking ensemble model is obtained, it will be used to classify the OCT strips in the test process. The results demonstrate that the proposed framework classifies overall OCT image strips and OCT strips containing MAs with accuracy scores of 0.982 and 0.987, respectively.

Purpose: To evaluate the performance of deep learning (DL) in diagnosing glaucoma and predicting its progression using fundus photography and retinal optical coherence tomography (OCT) images. Materials and Methods: Relevant studies published up to 30 October 2024 were retrieved from PubMed, Medline, EMBASE, Cochrane Library, Web of Science, and ClinicalKey. A bivariate random-effects model was employed to calculate pooled sensitivity, specificity, positive and negative likelihood ratios, and area under the receiver operating characteristic curve (AUROC). Results: A total of 48 studies were included in the meta-analysis. DL algorithms demonstrated high diagnostic performance in glaucoma detection using fundus photography and OCT images. For fundus photography, the pooled sensitivity and specificity were 0.92 (95% CI: 0.89-0.94) and 0.93 (95% CI: 0.90-0.95), respectively, with an AUROC of 0.90 (95% CI: 0.88-0.92). For the OCT imaging, the pooled sensitivity and specificity were 0.90 (95% CI: 0.84-0.94) and 0.87 (95% CI: 0.81-0.91), respectively, with an AUROC of 0.86 (95% CI: 0.83-0.90). In predicting glaucoma progression, DL models generally showed less robust performance, with pooled sensitivities and specificities ranging lower than in diagnostic tasks. Internal validation datasets showed higher accuracy than external validation datasets. Conclusions: DL algorithms achieve excellent performance in diagnosing glaucoma using fundus photography and OCT imaging. To enhance the prediction of glaucoma progression, future DL models should integrate multimodal data, including functional assessments, such as visual field measurements, and undergo extensive validation in real-world clinical settings.

PurposeTo explore clinical risk factors and optical coherence tomography (OCT) features associated with worse visual acuity (VA), progression of disease, choroidal neovascularization (CNV), and atrophy in eyes with adult-onset foveomacular vitelliform dystrophy (AOFVD). DesignSingle-center, retrospective, observational cohort study ParticipantsPatients seen at Duke Eye Center between January 2012 and May 2023 with a diagnosis of AOFVD confirmed via OCT and fundus autofluorescence. MethodsBaseline and final visit images from eyes with AOFVD were examined. Disease stage was assigned, and presence of atrophy or CNV was determined. Clinical and OCT features associated with progression to atrophy and CNV were determined using t-tests and chi-square analysis. Correlation with lower VA was determined using linear regression. Outcome MeasuresAssociation of clinical characteristics and OCT features with worse VA, progression of disease, CNV, and atrophy as determined by independent t-tests, chi-square analysis, and linear regression (p<0.05). Results101 eyes (63 patients) met inclusion criteria for this study, with mean follow-up duration of 48 months (SD 31 months). 51% of eyes progressed beyond baseline staging during follow-up; among baseline stage 1 eyes, incidence of atrophy was 0.068/person-year; incidence of CNV was 0.022/person-year. Risk factors for worse final VA were baseline presence of vitreomacular traction (VMT) (p=0.006), ellipsoid zone attenuation (p=0.024), increased lesion height and width (p<0.001). Predictors of progression include diabetes mellitus (p=0.012), statin use (p=0.031), presence of hyperreflective foci (p=0.012), and increased lesion width and volume (p=0.034, p=0.040). Predictors of atrophy include the baseline presence of VMT (p=0.018), decreased choroidal thickness (p=0.027), and greater maximal height, width, and volume of the lesion (p=0.031, p=0.020, p=0.009, respectively). Lower baseline VA (p=0.031) and increased lesion volume (p=0.042) were associated with CNV. ConclusionsClinical and OCT imaging features at baseline may prove useful in stratifying patient risk for progression, atrophy, CNV, and worse VA. Features such as statin use, diabetes, baseline VA, and laterality should be accounted for. OCT features such as lesion size, VMT, ellipsoid zone attenuation, choroidal thickness and hyperreflective foci may impart greater risk of poor outcomes. Future prospective analysis accounting for the time to development of atrophy and CNV is needed.

Collagen deposition disorders such as hypertrophic scars, keloids and scleroderma can be associated with significant stigma and embarrassment. These disorders often constitute considerable impairment to quality of life, with treatment posing to be a substantial challenge. Optical coherence tomography (OCT) provides a non-invasive, easily applicable bedside optical imaging method for assessment of the skin. It is hypothesized that OCT imaging may be useful in assessing fibrosis to avoid additional biopsies that could potentially worsen the scarring. Thirty-three patients with ordinary scars, hypertrophic scars, keloid scarring, lichen sclerosus et atrophicus and localized or systemic scleroderma were recruited for this pilot study. Affected tissue and adjacent healthy skin were scanned using OCT and digitally photographed. Density measurements were performed in ImageJ on OCT images from scleroderma patients, both systemic and morphea (10 patients), keloid patients (10 patients) and healthy skin adjacent to keloids (10 patients). OCT images of scarring diseases showed varying degrees of disruption to the skin architecture. OCT characteristics were identified for each lesion type. Hypertrophic scars displayed an increased vascularity and signal-rich bands correlating to excessive collagen deposition. Keloids depicted a disarray of hyper-reflective areas primarily located in the upper dermis. Additionally, the dermis displayed a heterogeneous morphology without indications of any vascular supply or lymphatic network. In contrast to keloids, scleroderma displayed a more cohesive backscattering indicating a difference in density of collagen or other dermal structures. OCT images demonstrated no significant differences between mean density measurements in OCT images of scleroderma, keloid and healthy skin (P=0.07). The OCT imaging appears to identify different scarring mechanisms, and therefore be of potential use in the assessment of outcomes following non-invasive therapy of e.g. early or progressive lesions.

To investigate morphological variations in the macular area with optical coherence tomography (OCT) after vitrectomy for diabetic fibrovascular proliferation. We reviewed 108 cases using OCT 7-15 months after vitrectomy. Of these, 32 received OCT within 3 months postoperatively. Morphological variations were categorized and correlated with visual outcome. Only 24 cases (21.4%) had no obvious abnormalities. The most frequent findings were epiretinal membrane (52.8%), macular thickening (37.0%) and macular cysts (28.7%). Multivariate regression showed that diffuse macular thickening, loss of foveal depression and diffuse retinal thinning were significantly associated with poor visual acuity. Sequential OCT (< 3 and >or= 7 months) revealed that epiretinal membrane and oedema outside of fovea changed significantly between two examinations. OCT may identify diverse morphological changes in the macular area after diabetic vitrectomy for fibrovascular proliferation. Macular appearance may change over time, and certain types of morphological changes may be associated with poor visual function.

Aim To develop an optimal method for automated segmentation of atherosclerotic plaque structural components in optical coherence tomography (OCT) images using an ensemble of deep learning neural network models based on a comparison of nine artificial neural network architectures.Material and methods This study utilized a multidisciplinary OCT dataset obtained from examinations of 103 patients. Each OCT session was annotated as a set of two-dimensional binary masks corresponding to the pixel boundaries of four key morphological features of plaques: vascular lumen, fibrous cap, lipid core, and microvessels. Nine deep learning models, including U-Net, DeepLabV3, and others, were used to segment the anatomical structures. Model hyperparameters were tuned using Bayesian optimization, and the performance was assessed using the Dice Similarity Coefficient (DSC) metric.Results The models demonstrated high segmentation accuracy for the vascular lumen (DSC: 0.987) and moderate accuracy for the fibrous cap (DSC: 0.736) and lipid core (DSC: 0.751). Microvessel segmentation proved more challenging, with a final DSC accuracy of 61%. A weighted ensemble of models, accounting for the uneven representation of different morphological structures and model confidence, achieved an average DSC of 88.2%, which significantly improved the overall accuracy compared to individual models. This accuracy value exceeds those of all individual models, including the highest DSC values of 0.784 (for microvessels) and 0.751 (for lipid core), indicating a consistent improvement in the segmentation accuracy by integrating the models.Conclusion The study confirms the effectiveness of the proposed strategy for segmentation of atherosclerotic plaques in OCT images. This strategy is based on using specialized models for various morphological features and a weighted ensemble adapted to the uneven representation of different morphological structures and morphological complexity. This approach provided a high segmentation accuracy despite the pronounced inequality in the representation of classes. The study results can contribute to the development of decision support methods in cardiology aimed at improving diagnostic accuracy and treatment of cardiovascular diseases.

Computer aided diagnosis of diabetic macular edema in retinal fundus and OCT images: A review