Effect of patch size and network architecture on a convolutional neural network approach for automatic segmentation of OCT retinal layers

Navigating the nuances: comparative analysis and hyperparameter optimisation of neural architectures on contrast-enhanced MRI for liver and liver tumour segmentation

In medical imaging, accurate segmentation is crucial to improving diagnosis, treatment, or both. However, navigating the multitude of available architectures for automatic segmentation can be overwhelming, making it challenging to determine the appropriate type of architecture and tune the most crucial parameters during dataset optimisation. To address this problem, we examined and refined seven distinct architectures for segmenting the liver, as well as liver tumours, with a restricted training collection of 60 3D contrast-enhanced magnetic resonance images (CE-MRI) from the ATLAS dataset. Included in these architectures are convolutional neural networks (CNNs), transformers, and hybrid CNN/transformer architectures. Bayesian search techniques were used for hyperparameter tuning to hasten convergence to the optimal parameter mixes while also minimising the number of trained models. It was unexpected that hybrid models, which typically exhibit superior performance on larger datasets, would exhibit comparable performance to CNNs. The optimisation of parameters contributed to better segmentations, resulting in an average increase of 1.7% and 5.0% in liver and tumour segmentation Dice coefficients, respectively. In conclusion, the findings of this study indicate that hybrid CNN/transformer architectures may serve as a practical substitute for CNNs even in small datasets. This underscores the significance of hyperparameter optimisation.

Optical Coherence Tomography-Derived Measurements of the Optic Nerve Head Structure of Aboriginal and Torres Strait Islander Children

PRCIS

This study demonstrated significant differences in optic nerve head characteristics in Aboriginal and Torres Strait Islander children compared with non-Indigenous children, which has implications for glaucoma risk and diagnosis in Aboriginal and Torres Strait Islander populations.

PURPOSE

The purpose of this study was to examine the optic nerve head (ONH) characteristics of visually normal Aboriginal and Torres Strait Islander children and non-Indigenous Australian children.

MATERIALS AND METHODS

Spectral domain optical coherence tomography imaging was performed on the right eye of 95 Aboriginal and Torres Strait Islander children and 149 non-Indigenous Australian children (5-18 years). Horizontal and vertical line scans, centered on the ONH, were analyzed to determine the dimensions of the ONH (Bruch membrane opening diameter), optic cup diameter, Bruch membrane opening minimum rim width, and the peripapillary retinal nerve fiber layer thickness.

RESULTS

The vertical but not horizontal Bruch membrane opening diameter of Aboriginal and Torres Strait Islander children was significantly larger than non-Indigenous children (mean difference: 0.09 mm, P = 0.001). The horizontal (mean difference: 0.12 mm, P = 0.003) and vertical cup diameter (mean difference: 0.16 mm, P < 0.001) were also significantly larger in Aboriginal and Torres Strait Islander children, as were the horizontal and vertical cup-to-disc ratios (both P < 0.01). Aboriginal and Torres Strait Islander children also had a significantly thinner Bruch membrane opening minimum rim width in the superior, nasal, and temporal meridians (all P < 0.001). Peripapillary retinal nerve fiber layer thickness did not differ between groups.

CONCLUSIONS

Differences exist in the ONH structure between Aboriginal and Torres Strait Islander children and non-Indigenous children, which may have implications for the detection and monitoring of ocular disease in this population and highlights the need to extend this research to the adult population.

Semantic segmentation of plant roots from RGB (mini-) rhizotron images-generalisation potential and false positives of established methods and advanced deep-learning models

BACKGROUND

Manual analysis of (mini-)rhizotron (MR) images is tedious. Several methods have been proposed for semantic root segmentation based on homogeneous, single-source MR datasets. Recent advances in deep learning (DL) have enabled automated feature extraction, but comparisons of segmentation accuracy, false positives and transferability are virtually lacking. Here we compare six state-of-the-art methods and propose two improved DL models for semantic root segmentation using a large MR dataset with and without augmented data. We determine the performance of the methods on a homogeneous maize dataset, and a mixed dataset of > 8 species (mixtures), 6 soil types and 4 imaging systems. The generalisation potential of the derived DL models is determined on a distinct, unseen dataset.

RESULTS

The best performance was achieved by the U-Net models; the more complex the encoder the better the accuracy and generalisation of the model. The heterogeneous mixed MR dataset was a particularly challenging for the non-U-Net techniques. Data augmentation enhanced model performance. We demonstrated the improved performance of deep meta-architectures and feature extractors, and a reduction in the number of false positives.

CONCLUSIONS

Although correction factors are still required to match human labelled root lengths, neural network architectures greatly reduce the time required to compute the root length. The more complex architectures illustrate how future improvements in root segmentation within MR images can be achieved, particularly reaching higher segmentation accuracies and model generalisation when analysing real-world datasets with artefacts-limiting the need for model retraining.

Automatic Segmentation of Retinal Layers in Multiple Neurodegenerative Disorder Scenarios

Retinal Optical Coherence Tomography (OCT) allows the non-invasive direct observation of the central nervous system, enabling the measurement and extraction of biomarkers from neural tissue that can be helpful in the assessment of ocular, systemic and Neurological Disorders (ND). Deep learning models can be trained to segment the retinal layers for biomarker extraction. However, the onset of ND can have an impact on the neural tissue, which can lead to the degraded performance of models not exposed to images displaying signs of disease during training. We present a fully automatic approach for the retinal layer segmentation in multiple neurodegenerative disorder scenarios, using an annotated dataset of patients of the most prevalent NDs: Alzheimer's disease, Parkinson's disease, multiple sclerosis and essential tremor, along with healthy control patients. Furthermore, we present a two-part, comprehensive study on the effects of ND on the performance of these models. The results show that images of healthy patients may not be sufficient for the robust training of automated segmentation models intended for the analysis of ND patients, and that using images representative of different NDs can increase the model performance. These results indicate that the presence or absence of patients of ND in datasets should be taken into account when training deep learning models for retinal layer segmentation, and that the proposed approach can provide a valuable tool for the robust and reliable diagnosis in multiple scenarios of ND.

Automatic Visual Acuity Loss Prediction in Children with Optic Pathway Gliomas using Magnetic Resonance Imaging

Children with optic pathway gliomas (OPGs), a low-grade brain tumor associated with neurofibromatosis type 1 (NF1-OPG), are at risk for permanent vision loss. While OPG size has been associated with vision loss, it is unclear how changes in size, shape, and imaging features of OPGs are associated with the likelihood of vision loss. This paper presents a fully automatic framework for accurate prediction of visual acuity loss using multi-sequence magnetic resonance images (MRIs). Our proposed framework includes a transformer-based segmentation network using transfer learning, statistical analysis of radiomic features, and a machine learning method for predicting vision loss. Our segmentation network was evaluated on multi-sequence MRIs acquired from 75 pediatric subjects with NF1-OPG and obtained an average Dice similarity coefficient of 0.791. The ability to predict vision loss was evaluated on a subset of 25 subjects with ground truth using cross-validation and achieved an average accuracy of 0.8. Analyzing multiple MRI features appear to be good indicators of vision loss, potentially permitting early treatment decisions.Clinical relevance- Accurately determining which children with NF1-OPGs are at risk and hence require preventive treatment before vision loss remains challenging, towards this we present a fully automatic deep learning-based framework for vision outcome prediction, potentially permitting early treatment decisions.

Enhanced Deep Learning Model for Classification of Retinal Optical Coherence Tomography Images

Retinal optical coherence tomography (OCT) imaging is a valuable tool for assessing the condition of the back part of the eye. The condition has a great effect on the specificity of diagnosis, the monitoring of many physiological and pathological procedures, and the response and evaluation of therapeutic effectiveness in various fields of clinical practices, including primary eye diseases and systemic diseases such as diabetes. Therefore, precise diagnosis, classification, and automated image analysis models are crucial. In this paper, we propose an enhanced optical coherence tomography (EOCT) model to classify retinal OCT based on modified ResNet (50) and random forest algorithms, which are used in the proposed study's training strategy to enhance performance. The Adam optimizer is applied during the training process to increase the efficiency of the ResNet (50) model compared with the common pre-trained models, such as spatial separable convolutions and visual geometry group (VGG) (16). The experimentation results show that the sensitivity, specificity, precision, negative predictive value, false discovery rate, false negative rate accuracy, and Matthew's correlation coefficient are 0.9836, 0.9615, 0.9740, 0.9756, 0.0385, 0.0260, 0.0164, 0.9747, 0.9788, and 0.9474, respectively.

Multiscale joint segmentation method for retinal optical coherence tomography images using a bidirectional wave algorithm and improved graph theory

Morphology and functional metrics of retinal layers are important biomarkers for many human ophthalmic diseases. Automatic and accurate segmentation of retinal layers is crucial for disease diagnosis and research. To improve the performance of retinal layer segmentation, a multiscale joint segmentation framework for retinal optical coherence tomography (OCT) images based on bidirectional wave algorithm and improved graph theory is proposed. In this framework, the bidirectional wave algorithm was used to segment edge information in multiscale images, and the improved graph theory was used to modify edge information globally, to realize automatic and accurate segmentation of eight retinal layer boundaries. This framework was tested on two public datasets and two OCT imaging systems. The test results show that, compared with other state-of-the-art methods, this framework does not need data pre-training and parameter pre-adjustment on different datasets, and can achieve sub-pixel retinal layer segmentation on a low-configuration computer.

AI-based MRI auto-segmentation of brain tumor in rodents, a multicenter study

Automatic segmentation of rodent brain tumor on magnetic resonance imaging (MRI) may facilitate biomedical research. The current study aims to prove the feasibility for automatic segmentation by artificial intelligence (AI), and practicability of AI-assisted segmentation. MRI images, including T2WI, T1WI and CE-T1WI, of brain tumor from 57 WAG/Rij rats in KU Leuven and 46 mice from the cancer imaging archive (TCIA) were collected. A 3D U-Net architecture was adopted for segmentation of tumor bearing brain and brain tumor. After training, these models were tested with both datasets after Gaussian noise addition. Reduction of inter-observer disparity by AI-assisted segmentation was also evaluated. The AI model segmented tumor-bearing brain well for both Leuven and TCIA datasets, with Dice similarity coefficients (DSCs) of 0.87 and 0.85 respectively. After noise addition, the performance remained unchanged when the signal-noise ratio (SNR) was higher than two or eight, respectively. For the segmentation of tumor lesions, AI-based model yielded DSCs of 0.70 and 0.61 for Leuven and TCIA datasets respectively. Similarly, the performance is uncompromised when the SNR was over two and eight respectively. AI-assisted segmentation could significantly reduce the inter-observer disparities and segmentation time in both rats and mice. Both AI models for segmenting brain or tumor lesions could improve inter-observer agreement and therefore contributed to the standardization of the following biomedical studies.

Patch-based CNN for corneal segmentation of AS-OCT images: Effect of the number of classes and image quality upon performance

Anterior segment optical coherence tomography (AS-OCT) is a fundamental ophthalmic imaging technique. AS-OCT images can be examined by experts and segmented to provide quantitative metrics that inform clinical decision making. Manual segmentation of these images is time-consuming and subjective, encouraging software developers in the field to automate segmentation procedures. Traditional programing segmentation approaches are being replaced by deep learning methods, which have shown state-of-the-art performance in AS-OCT image analysis. In this study, a method based on patch-based convolutional neural networks (CNN) was used to segment the three main boundaries of the cornea: the epithelium, Bowman's layer, and the endothelium. To assess the effect of the number of classes on performance, the model was designed as a patch-based boundary classifier using 4 and 8 classes. The effect of image quality was also assessed using different data distributions during the training process. While the Dice coefficient and probability revealed greater precision for the 8 class models, the boundary error metric indicated comparable performance. Additionally, for 8 class models, the image quality test had only a small negative effect on performance, which may be an indication of the robustness of the model and could also suggest that the data augmentation methods did not show significant improvement. These findings contribute to the development of automatic segmentation techniques for AS-OCT images, since patch-based methods have been largely unexplored in favor of other deep learning techniques. The overall performance of the proposed method is comparable to other well-established segmentation methods.

FAPNET: Feature Fusion with Adaptive Patch for Flood-Water Detection and Monitoring

In satellite remote sensing applications, waterbody segmentation plays an essential role in mapping and monitoring the dynamics of surface water. Satellite image segmentation-examining a relevant sensor data spectrum and identifying the regions of interests to obtain improved performance-is a fundamental step in satellite data analytics. Satellite image segmentation is challenging for a number of reasons, which include cloud interference, inadequate label data, low lighting and the presence of terrain. In recent years, Convolutional Neural Networks (CNNs), combined with (satellite captured) multispectral image segmentation techniques, have led to promising advances in related research. However, ensuring sufficient image resolution, maintaining class balance to achieve prediction quality and reducing the computational overhead of the deep neural architecture are still open to research due to the sophisticated CNN hierarchical architectures. To address these issues, we propose a number of methods: a multi-channel Data-Fusion Module (DFM), Neural Adaptive Patch (NAP) augmentation algorithm and re-weight class balancing (implemented in our PHR-CB experimental setup). We integrated these techniques into our novel Fusion Adaptive Patch Network (FAPNET). Our dataset is the Sentinel-1 SAR microwave signal, used in the Microsoft Artificial Intelligence for Earth competition, so that we can compare our results with the top scores in the competition. In order to validate our approach, we designed four experimental setups and in each setup, we compared our results with the popular image segmentation models UNET, VNET, DNCNN, UNET++, U2NET, ATTUNET, FPN and LINKNET. The comparisons demonstrate that our PHR-CB setup, with class balance, generates the best performance for all models in general and our FAPNET approach outperforms relative works. FAPNET successfully detected the salient features from the satellite images. FAPNET with a MeanIoU score of 87.06% outperforms the state-of-the-art UNET, which has a score of 79.54%. In addition, FAPNET has a shorter training time than other models, comparable to that of UNET (6.77 min for 5 epochs). Qualitative analysis also reveals that our FAPNET model successfully distinguishes micro waterbodies better than existing models. FAPNET is more robust to low lighting, cloud and weather fluctuations and can also be used in RGB images. Our proposed method is lightweight, computationally inexpensive, robust and simple to deploy in industrial applications. Our research findings show that flood-water mapping is more accurate when using SAR signals than RGB images. Our FAPNET architecture, having less parameters than UNET, can distinguish micro waterbodies accurately with shorter training time.

Gradual Self-Training via Confidence and Volume Based Domain Adaptation for Multi Dataset Deep Learning-Based Brain Metastases Detection Using Nonlocal Networks on MRI Images

BACKGROUND

Research suggests that treatment of multiple brain metastases (BMs) with stereotactic radiosurgery shows improvement when metastases are detected early, providing a case for BM detection capabilities on small lesions.

PURPOSE

To demonstrate automatic detection of BM on three MRI datasets using a deep learning-based approach. To improve the performance of the network is iteratively co-trained with datasets from different domains. A systematic approach is proposed to prevent catastrophic forgetting during co-training.

STUDY TYPE

Retrospective.

POPULATION

A total of 156 patients (105 ground truth and 51 pseudo labels) with 1502 BM (BrainMetShare); 121 patients with 722 BM (local); 400 patients with 447 primary gliomas (BrATS). Training/pseudo labels/validation data were distributed 84/51/21 (BrainMetShare). Training/validation data were split: 121/23 (local) and 375/25 (BrATS).

FIELD STRENGTH/SEQUENCE

A 5 T and 3 T/T1 spin-echo postcontrast (T1-gradient echo) (BrainMetShare), 3 T/T1 magnetization prepared rapid acquisition gradient echo postcontrast (T1-MPRAGE) (local), 0.5 T, 1 T, and 1.16 T/T1-weighted-fluid-attenuated inversion recovery (T1-FLAIR) (BrATS).

ASSESSMENT

The ground truth was manually segmented by two (BrainMetShare) and four (BrATS) radiologists and manually annotated by one (local) radiologist. Confidence and volume based domain adaptation (CAVEAT) method of co-training the three datasets on a 3D nonlocal convolutional neural network (CNN) architecture was implemented to detect BM.

STATISTICAL TESTS

The performance was evaluated using sensitivity and false positive rates per patient (FP/patient) and free receiver operating characteristic (FROC) analysis at seven predefined (1/8, 1/4, 1/2, 1, 2, 4, and 8) FPs per scan.

RESULTS

The sensitivity and FP/patient from a held-out set registered 0.811 at 2.952 FP/patient (BrainMetShare), 0.74 at 3.130 (local), and 0.723 at 2.240 (BrATS) using the CAVEAT approach with lesions as small as 1 mm being detected.

DATA CONCLUSION

Improved sensitivities at lower FP can be achieved by co-training datasets via the CAVEAT paradigm to address the problem of data sparsity.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY STAGE: 2.

A comparison of deep learning U-Net architectures for posterior segment OCT retinal layer segmentation

Deep learning methods have enabled a fast, accurate and automated approach for retinal layer segmentation in posterior segment OCT images. Due to the success of semantic segmentation methods adopting the U-Net, a wide range of variants and improvements have been developed and applied to OCT segmentation. Unfortunately, the relative performance of these methods is difficult to ascertain for OCT retinal layer segmentation due to a lack of comprehensive comparative studies, and a lack of proper matching between networks in previous comparisons, as well as the use of different OCT datasets between studies. In this paper, a detailed and unbiased comparison is performed between eight U-Net architecture variants across four different OCT datasets from a range of different populations, ocular pathologies, acquisition parameters, instruments and segmentation tasks. The U-Net architecture variants evaluated include some which have not been previously explored for OCT segmentation. Using the Dice coefficient to evaluate segmentation performance, minimal differences were noted between most of the tested architectures across the four datasets. Using an extra convolutional layer per pooling block gave a small improvement in segmentation performance for all architectures across all four datasets. This finding highlights the importance of careful architecture comparison (e.g. ensuring networks are matched using an equivalent number of layers) to obtain a true and unbiased performance assessment of fully semantic models. Overall, this study demonstrates that the vanilla U-Net is sufficient for OCT retinal layer segmentation and that state-of-the-art methods and other architectural changes are potentially unnecessary for this particular task, especially given the associated increased complexity and slower speed for the marginal performance gains observed. Given the U-Net model and its variants represent one of the most commonly applied image segmentation methods, the consistent findings across several datasets here are likely to translate to many other OCT datasets and studies. This will provide significant value by saving time and cost in experimentation and model development as well as reduced inference time in practice by selecting simpler models.

Retinal thickness in healthy Australian Aboriginal and Torres Strait Islander children

Background

Understanding normative retinal thickness characteristics is critical for diagnosis and monitoring of pathology, particularly in those predisposed to retinal disease. The macular retinal layer thickness of Australian Aboriginal and/or Torres Strait Islander children was examined using spectral-domain optical coherence tomography.

Methods

High-resolution macular optical coherence tomography imaging was performed on 100 Aboriginal and/or Torres Strait Islander children and 150 non-Indigenous visually healthy children aged 4–18 years. The imaging protocol included a 6-line radial scan centred on the fovea. Images were segmented using semi-automated software to derive thickness of the total retina, inner and outer retina, and individual retinal layers across the macular region. Repeated measures ANOVAs examined variations in thickness associated with retinal region, age, gender and Indigenous status.

Results

Retinal thickness showed significant topographical variations (p < 0.01), being thinnest in the foveal zone, and thickest in the parafovea. The retina of Aboriginal and/or Torres Strait Islander children was significantly thinner than non-Indigenous children in the foveal (p < 0.001), parafoveal (p = 0.002), and perifoveal zones (p = 0.01), with the greatest difference in the foveal zone (mean difference: 14.2 μm). Inner retinal thickness was also thinner in Aboriginal and/or Torres Strait Islander children compared to non-Indigenous children in the parafoveal zone (p < 0.001), and outer retinal thickness was thinner in the foveal (p < 0.001) and perifoveal zone (p < 0.001). Retinal thickness was also significantly greater in males than females (p < 0.001) and showed a statistically significant positive association with age (p = 0.01).

Conclusion

There are significant differences in macular retinal thickness between Aboriginal and/or Torres Strait Islander children and non-Indigenous children, which has implications for interpreting optical coherence tomography data and may relate to risk of macula disease in this population.

Deep-XFCT: Deep Learning 3D-Mineral Liberation Analysis with Micro-X-ray Fluorescence and Computed Tomography

Quantitative characterisation through mineral liberation analysis is required for effective minerals processing in areas such as mineral deposits, tailings and reservoirs in industries for resources, environment and materials science. Current practices in mineral liberation analysis are based on 2D representations, leading to systematic errors in the extrapolation to 3D volumetric properties. The rapid development of X-ray microcomputed tomography (μCT) opens new opportunities for 3D analysis of features such as particle- and grain-size characterisation, determination of particle densities and shape factors, estimation of mineral associations, and liberation and locking. To date, no simple non-destructive method exists for 3D mineral liberation analysis. We present a new development based on combining μCT with micro-X-ray fluorescence (μXRF) using deep learning. We demonstrate successful semi-automated multimodal analysis of a crystalline magmatic rock by obtaining 2D μXRF mineral maps from the top and bottom of the cylindrical core and propagating that information through the 3D μCT volume with deep learning segmentation. The deep learning model was able to segment the core to obtain reasonable mineral attributes. Additionally, the model overcame the challenge of differentiating minerals with similar densities in μCT, which would not be possible with conventional segmentation methods. The approach is universal and can be extended to any multimodal and multi-instrument analysis for further refinement. We conclude that the combination of μCT and μXRF can provide a new opportunity for robust 3D mineral liberation analysis in both field and laboratory applications.

Intraretinal Layer Segmentation Using Cascaded Compressed U-Nets

Reliable biomarkers quantifying neurodegeneration and neuroinflammation in central nervous system disorders such as Multiple Sclerosis, Alzheimer's dementia or Parkinson's disease are an unmet clinical need. Intraretinal layer thicknesses on macular optical coherence tomography (OCT) images are promising noninvasive biomarkers querying neuroretinal structures with near cellular resolution. However, changes are typically subtle, while tissue gradients can be weak, making intraretinal segmentation a challenging task. A robust and efficient method that requires no or minimal manual correction is an unmet need to foster reliable and reproducible research as well as clinical application. Here, we propose and validate a cascaded two-stage network for intraretinal layer segmentation, with both networks being compressed versions of U-Net (CCU-INSEG). The first network is responsible for retinal tissue segmentation from OCT B-scans. The second network segments eight intraretinal layers with high fidelity. At the post-processing stage, we introduce Laplacian-based outlier detection with layer surface hole filling by adaptive non-linear interpolation. Additionally, we propose a weighted version of focal loss to minimize the foreground-background pixel imbalance in the training data. We train our method using 17,458 B-scans from patients with autoimmune optic neuropathies, i.e., multiple sclerosis, and healthy controls. Voxel-wise comparison against manual segmentation produces a mean absolute error of 2.3 μm, outperforming current state-of-the-art methods on the same data set. Voxel-wise comparison against external glaucoma data leads to a mean absolute error of 2.6 μm when using the same gold standard segmentation approach, and 3.7 μm mean absolute error in an externally segmented data set. In scans from patients with severe optic atrophy, 3.5% of B-scan segmentation results were rejected by an experienced grader, whereas this was the case in 41.4% of B-scans segmented with a graph-based reference method. The validation results suggest that the proposed method can robustly segment macular scans from eyes with even severe neuroretinal changes.

Automated 2D Slice-Based Skull Stripping Multi-View Ensemble Model on NFBS and IBSR Datasets

This study proposed and evaluated a two-dimensional (2D) slice-based multi-view U-Net (MVU-Net) architecture for skull stripping. The proposed model fused all three TI-weighted brain magnetic resonance imaging (MRI) views, i.e., axial, coronal, and sagittal. This 2D method performed equally well as a three-dimensional (3D) model of skull stripping. while using fewer computational resources. The predictions of all three views were fused linearly, producing a final brain mask with better accuracy and efficiency. Meanwhile, two publicly available datasets-the Internet Brain Segmentation Repository (IBSR) and Neurofeedback Skull-stripped (NFBS) repository-were trained and tested. The MVU-Net, U-Net, and skip connection U-Net (SCU-Net) architectures were then compared. For the IBSR dataset, compared to U-Net and SC-UNet, the MVU-Net architecture attained better mean dice score coefficient (DSC), sensitivity, and specificity, at 0.9184, 0.9397, and 0.9908, respectively. Similarly, the MVU-Net architecture achieved better mean DSC, sensitivity, and specificity, at 0.9681, 0.9763, and 0.9954, respectively, than the U-Net and SC-UNet for the NFBS dataset.

Artificial Intelligence as a Tool to Study the 3D Skeletal Architecture in Newly Settled Coral Recruits: Insights into the Effects of Ocean Acidification on Coral Biomineralization

Understanding the formation of the coral skeleton has been a common subject uniting various marine and materials study fields. Two main regions dominate coral skeleton growth: Rapid Accretion Deposits (RADs) and Thickening Deposits (TDs). These have been extensively characterized at the 2D level, but their 3D characteristics are still poorly described. Here, we present an innovative approach to combine synchrotron phase contrast-enhanced microCT (PCE-CT) with artificial intelligence (AI) to explore the 3D architecture of RADs and TDs within the coral skeleton. As a reference study system, we used recruits of the stony coral Stylophora pistillata from the Red Sea, grown under both natural and simulated ocean acidification conditions. We thus studied the recruit’s skeleton under both regular and morphologically-altered acidic conditions. By imaging the corals with PCE-CT, we revealed the interwoven morphologies of RADs and TDs. Deep-learning neural networks were invoked to explore AI segmentation of these regions, to overcome limitations of common segmentation techniques. This analysis yielded highly-detailed 3D information about the RAD’s and TD’s architecture. Our results demonstrate how AI can be used as a powerful tool to obtain 3D data essential for studying coral biomineralization and for exploring the effects of environmental change on coral growth.

OCT Retinal and Choroidal Layer Instance Segmentation Using Mask R-CNN

Optical coherence tomography (OCT) of the posterior segment of the eye provides high-resolution cross-sectional images that allow visualization of individual layers of the posterior eye tissue (the retina and choroid), facilitating the diagnosis and monitoring of ocular diseases and abnormalities. The manual analysis of retinal OCT images is a time-consuming task; therefore, the development of automatic image analysis methods is important for both research and clinical applications. In recent years, deep learning methods have emerged as an alternative method to perform this segmentation task. A large number of the proposed segmentation methods in the literature focus on the use of encoder–decoder architectures, such as U-Net, while other architectural modalities have not received as much attention. In this study, the application of an instance segmentation method based on region proposal architecture, called the Mask R-CNN, is explored in depth in the context of retinal OCT image segmentation. The importance of adequate hyper-parameter selection is examined, and the performance is compared with commonly used techniques. The Mask R-CNN provides a suitable method for the segmentation of OCT images with low segmentation boundary errors and high Dice coefficients, with segmentation performance comparable with the commonly used U-Net method. The Mask R-CNN has the advantage of a simpler extraction of the boundary positions, especially avoiding the need for a time-consuming graph search method to extract boundaries, which reduces the inference time by 2.5 times compared to U-Net, while segmenting seven retinal layers.

Application of Deep Learning Methods for Binarization of the Choroid in Optical Coherence Tomography Images

Purpose

The purpose of this study was to develop a deep learning model for automatic binarization of the choroidal tissue, separating choroidal blood vessels from nonvascular stromal tissue, in optical coherence tomography (OCT) images from healthy young subjects.

Methods

OCT images from an observational longitudinal study of 100 children were used for training, validation, and testing of 5 fully semantic networks, which provided a binarized output of the choroid. These outputs were compared with ground truth images, generated from a local binarization technique after manually optimizing the analysis window size for each individual image. The performance was evaluated using accuracy and repeatability metrics. The methods were also compared with a fixed window size local binarization technique, which has been commonly used previously.

Results

The tested deep learning methods provided a good performance in terms of accuracy and repeatability. With the U-Net and SegNet networks showing >96% accuracy. All methods displayed a high level of repeatability relative to the ground truth. For analysis of the choroidal vascularity index (a commonly used metric derived from the binarized image), SegNet showed the closest agreement with the ground truth and high repeatability. The fixed window size showed a reduced accuracy compared to other methods.

Conclusions

Fully semantic networks such as U-Net and SegNet displayed excellent performance for the binarization task. These methods provide a useful approach for clinical and research applications of deep learning tools for the binarization of the choroid in OCT images.

Translational Relevance

Deep learning models provide a novel, robust solution to automatically binarize the choroidal tissue in OCT images.

Automated retinal boundary segmentation of optical coherence tomography images using an improved Canny operator

Retinal segmentation is a prerequisite for quantifying retinal structural features and diagnosing related ophthalmic diseases. Canny operator is recognized as the best boundary detection operator so far, and is often used to obtain the initial boundary of the retina in retinal segmentation. However, the traditional Canny operator is susceptible to vascular shadows, vitreous artifacts, or noise interference in retinal segmentation, causing serious misdetection or missed detection. This paper proposed an improved Canny operator for automatic segmentation of retinal boundaries. The improved algorithm solves the problems of the traditional Canny operator by adding a multi-point boundary search step on the basis of the original method, and adjusts the convolution kernel. The algorithm was used to segment the retinal images of healthy subjects and age-related macular degeneration (AMD) patients; eleven retinal boundaries were identified and compared with the results of manual segmentation by the ophthalmologists. The average difference between the automatic and manual methods is: 2–6 microns (1–2 pixels) for healthy subjects and 3–10 microns (1–3 pixels) for AMD patients. Qualitative method is also used to verify the accuracy and stability of the algorithm. The percentage of “perfect segmentation” and “good segmentation” is 98% in healthy subjects and 94% in AMD patients. This algorithm can be used alone or in combination with other methods as an initial boundary detection algorithm. It is easy to understand and improve, and may become a useful tool for analyzing and diagnosing eye diseases.

floodGAN: Using Deep Adversarial Learning to Predict Pluvial Flooding in Real Time

Using machine learning for pluvial flood prediction tasks has gained growing attention in the past years. In particular, data-driven models using artificial neuronal networks show promising results, shortening the computation times of physically based simulations. However, recent approaches have used mainly conventional fully connected neural networks which were (a) restricted to spatially uniform precipitation events and (b) limited to a small amount of input data. In this work, a deep convolutional generative adversarial network has been developed to predict pluvial flooding caused by nonlinear spatial heterogeny rainfall events. The model developed, floodGAN, is based on an image-to-image translation approach whereby the model learns to generate 2D inundation predictions conditioned by heterogenous rainfall distributions—through the minimax game of two adversarial networks. The training data for the floodGAN model was generated using a physically based hydrodynamic model. To evaluate the performance and accuracy of the floodGAN, model multiple tests were conducted using both synthetic events and a historic rainfall event. The results demonstrate that the proposed floodGAN model is up to 106 times faster than the hydrodynamic model and promising in terms of accuracy and generalizability. Therefore, it bridges the gap between detailed flood modelling and real-time applications such as end-to-end early warning systems.

Evaluation of focus and deep learning methods for automated image grading and factors influencing image quality in adaptive optics ophthalmoscopy

Adaptive optics flood illumination ophthalmoscopy (AO-FIO) is an established imaging tool in the investigation of retinal diseases. However, the clinical interpretation of AO-FIO images can be challenging due to varied image quality. Therefore, image quality assessment is essential before interpretation. An image assessment tool will also assist further work on improving the image quality, either during acquisition or post processing. In this paper, we describe, validate and compare two automated image quality assessment methods; the energy of Laplacian focus operator (LAPE; not commonly used but easily implemented) and convolutional neural network (CNN; effective but more complex approach). We also evaluate the effects of subject age, axial length, refractive error, fixation stability, disease status and retinal location on AO-FIO image quality. Based on analysis of 10,250 images of 50 × 50 μm size, at 41 retinal locations, from 50 subjects we demonstrate that CNN slightly outperforms LAPE in image quality assessment. CNN achieves accuracy of 89%, whereas LAPE metric achieves 73% and 80% (for a linear regression and random forest multiclass classifier methods, respectively) compared to ground truth. Furthermore, the retinal location, age and disease are factors that can influence the likelihood of poor image quality.

Corneal pachymetry by AS-OCT after Descemet's membrane endothelial keratoplasty

Corneal thickness (pachymetry) maps can be used to monitor restoration of corneal endothelial function, for example after Descemet’s membrane endothelial keratoplasty (DMEK). Automated delineation of the corneal interfaces in anterior segment optical coherence tomography (AS-OCT) can be challenging for corneas that are irregularly shaped due to pathology, or as a consequence of surgery, leading to incorrect thickness measurements. In this research, deep learning is used to automatically delineate the corneal interfaces and measure corneal thickness with high accuracy in post-DMEK AS-OCT B-scans. Three different deep learning strategies were developed based on 960 B-scans from 50 patients. On an independent test set of 320 B-scans, corneal thickness could be measured with an error of 13.98 to 15.50 μm for the central 9 mm range, which is less than 3% of the average corneal thickness. The accurate thickness measurements were used to construct detailed pachymetry maps. Moreover, follow-up scans could be registered based on anatomical landmarks to obtain differential pachymetry maps. These maps may enable a more comprehensive understanding of the restoration of the endothelial function after DMEK, where thickness often varies throughout different regions of the cornea, and subsequently contribute to a standardized postoperative regime.

Glaucoma classification based on scanning laser ophthalmoscopic images using a deep learning ensemble method

This study aimed to assess the utility of optic nerve head (onh) en-face images, captured with scanning laser ophthalmoscopy (slo) during standard optical coherence tomography (oct) imaging of the posterior segment, and demonstrate the potential of deep learning (dl) ensemble method that operates in a low data regime to differentiate glaucoma patients from healthy controls. The two groups of subjects were initially categorized based on a range of clinical tests including measurements of intraocular pressure, visual fields, oct derived retinal nerve fiber layer (rnfl) thickness and dilated stereoscopic examination of onh. 227 slo images of 227 subjects (105 glaucoma patients and 122 controls) were used. A new task-specific convolutional neural network architecture was developed for slo image-based classification. To benchmark the results of the proposed method, a range of classifiers were tested including five machine learning methods to classify glaucoma based on rnfl thickness—a well-known biomarker in glaucoma diagnostics, ensemble classifier based on inception v3 architecture, and classifiers based on features extracted from the image. The study shows that cross-validation dl ensemble based on slo images achieved a good discrimination performance with up to 0.962 of balanced accuracy, outperforming all of the other tested classifiers.

A Lightweight Depth Estimation Network for Wide-Baseline Light Fields

Existing traditional and ConvNet-based methods for light field depth estimation mainly work on the narrow-baseline scenario. This paper explores the feasibility and capability of ConvNets to estimate depth in another promising scenario: wide-baseline light fields. Due to the deficiency of training samples, a large-scale and diverse synthetic wide-baseline dataset with labelled data is introduced for depth prediction tasks. Considering the practical goal for real-world applications, we design an end-to-end trained lightweight convolutional network to infer depths from light fields, called LLF-Net. The proposed LLF-Net is built by incorporating a cost volume which allows variable angular light field inputs and an attention module that enables to recover details at occlusion areas. Evaluations are made on the synthetic and real-world wide-baseline light fields, and experimental results show that the proposed network achieves the best performance when compared to recent state-of-the-art methods. We also evaluate our LLF-Net on narrow-baseline datasets, and it consequently improves the performance of previous methods.

DeepRetina: Layer Segmentation of Retina in OCT Images Using Deep Learning

Purpose

To automate the segmentation of retinal layers, we propose DeepRetina, a method based on deep neural networks.

Methods

DeepRetina uses the improved Xception65 to extract and learn the characteristics of retinal layers. The Xception65-extracted feature maps are inputted to an atrous spatial pyramid pooling module to obtain multiscale feature information. This information is then recovered to capture clearer retinal layer boundaries in the encoder-decoder module, thus completing retinal layer auto-segmentation of the retinal optical coherence tomography (OCT) images.

Results

We validated this method using a retinal OCT image database containing 280 volumes (40 B-scans per volume) to demonstrate its effectiveness. The results showed that the method exhibits excellent performance in terms of the mean intersection over union and sensitivity (Se), which are as high as 90.41 and 92.15%, respectively. The intersection over union and Se values of the nerve fiber layer, ganglion cell layer, inner plexiform layer, inner nuclear layer, outer plexiform layer, outer nuclear layer, outer limiting membrane, photoreceptor inner segment, photoreceptor outer segment, and pigment epithelium layer were found to be above 88%.

Conclusions

DeepRetina can automate the segmentation of retinal layers and has great potential for the early diagnosis of fundus retinal diseases. In addition, our approach will provide a segmentation model framework for other types of tissues and cells in clinical practice.

Translational Relevance

Automating the segmentation of retinal layers can help effectively diagnose and monitor clinical retinal diseases. In addition, it requires only a small amount of manual segmentation, significantly improving work efficiency.

Retinal Boundary Segmentation in Stargardt Disease Optical Coherence Tomography Images Using Automated Deep Learning

Purpose

To use a deep learning model to develop a fully automated method (fully semantic network and graph search [FS-GS]) of retinal segmentation for optical coherence tomography (OCT) images from patients with Stargardt disease.

Methods

Eighty-seven manually segmented (ground truth) OCT volume scan sets (5171 B-scans) from 22 patients with Stargardt disease were used for training, validation and testing of a novel retinal boundary detection approach (FS-GS) that combines a fully semantic deep learning segmentation method, which generates a per-pixel class prediction map with a graph-search method to extract retinal boundary positions. The performance was evaluated using the mean absolute boundary error and the differences in two clinical metrics (retinal thickness and volume) compared with the ground truth. The performance of a separate deep learning method and two publicly available software algorithms were also evaluated against the ground truth.

Results

FS-GS showed an excellent agreement with the ground truth, with a boundary mean absolute error of 0.23 and 1.12 pixels for the internal limiting membrane and the base of retinal pigment epithelium or Bruch's membrane, respectively. The mean difference in thickness and volume across the central 6 mm zone were 2.10 µm and 0.059 mm³. The performance of the proposed method was more accurate and consistent than the publicly available OCTExplorer and AURA tools.

Conclusions

The FS-GS method delivers good performance in segmentation of OCT images of pathologic retina in Stargardt disease.

Translational Relevance

Deep learning models can provide a robust method for retinal segmentation and support a high-throughput analysis pipeline for measuring retinal thickness and volume in Stargardt disease.

Approaches to quantify optical coherence tomography angiography metrics

Optical coherence tomography (OCT) has revolutionized the field of ophthalmology in the last three decades. As an OCT extension, OCT angiography (OCTA) utilizes a fast OCT system to detect motion contrast in ocular tissue and provides a three-dimensional representation of the ocular vasculature in a non-invasive, dye-free manner. The first OCT machine equipped with OCTA function was approved by U.S. Food and Drug Administration in 2016 and now it is widely applied in clinics. To date, numerous methods have been developed to aid OCTA interpretation and quantification. In this review, we focused on the workflow of OCTA-based interpretation, beginning from the generation of the OCTA images using signal decorrelation, which we divided into intensity-based, phase-based and phasor-based methods. We further discussed methods used to address image artifacts that are commonly observed in clinical settings, to the algorithms for image enhancement, binarization, and OCTA metrics extraction. We believe a better grasp of these technical aspects of OCTA will enhance the understanding of the technology and its potential application in disease diagnosis and management. Moreover, future studies will also explore the use of ocular OCTA as a window to link ocular vasculature to the function of other organs such as the kidney and brain.

Application of Artificial Intelligence in Targeting Retinal Diseases

Retinal diseases affect an increasing number of patients worldwide because of the aging population. Request for diagnostic imaging in ophthalmology is ramping up, while the number of specialists keeps shrinking. Cutting-edge technology embedding artificial intelligence (AI) algorithms are thus advocated to help ophthalmologists perform their clinical tasks as well as to provide a source for the advancement of novel biomarkers. In particular, optical coherence tomography (OCT) evaluation of the retina can be augmented by algorithms based on machine learning and deep learning to early detect, qualitatively localize and quantitatively measure epi/intra/subretinal abnormalities or pathological features of macular or neural diseases. In this paper, we discuss the use of AI to facilitate efficacy and accuracy of retinal imaging in those diseases increasingly treated by intravitreal vascular endothelial growth factor (VEGF) inhibitors (i.e. anti-VEGF drugs), also including integration and interpretation features in the process. We review recent advances by AI in diabetic retinopathy, age-related macular degeneration, and retinopathy of prematurity that envision a potentially key role of highly automated systems in screening, early diagnosis, grading and individualized therapy. We discuss benefits and critical aspects of automating the evaluation of disease activity, recurrences, the timing of retreatment and therapeutically potential novel targets in ophthalmology. The impact of massive employment of AI to optimize clinical assistance and encourage tailored therapies for distinct patterns of retinal diseases is also discussed.

Adversarial convolutional network for esophageal tissue segmentation on OCT images

Automatic segmentation is important for esophageal OCT image processing, which is able to provide tissue characteristics such as shape and thickness for disease diagnosis. Existing automatical segmentation methods based on deep convolutional networks may not generate accurate segmentation results due to limited training set and various layer shapes. This study proposed a novel adversarial convolutional network (ACN) to segment esophageal OCT images using a convolutional network trained by adversarial learning. The proposed framework includes a generator and a discriminator, both with U-Net alike fully convolutional architecture. The discriminator is a hybrid network that discriminates whether the generated results are real and implements pixel classification at the same time. Leveraging on the adversarial training, the discriminator becomes more powerful. In addition, the adversarial loss is able to encode high order relationships of pixels, thus eliminating the requirements of post-processing. Experiments on segmenting esophageal OCT images from guinea pigs confirmed that the ACN outperforms several deep learning frameworks in pixel classification accuracy and improves the segmentation result. The potential clinical application of ACN for detecting eosinophilic esophagitis (EoE), an esophageal disease, is also presented in the experiment.

CT Super-Resolution GAN Constrained by the Identical, Residual, and Cycle Learning Ensemble (GAN-CIRCLE)

In this paper, we present a semi-supervised deep learning approach to accurately recover high-resolution (HR) CT images from low-resolution (LR) counterparts. Specifically, with the generative adversarial network (GAN) as the building block, we enforce the cycle-consistency in terms of the Wasserstein distance to establish a nonlinear end-to-end mapping from noisy LR input images to denoised and deblurred HR outputs. We also include the joint constraints in the loss function to facilitate structural preservation. In this process, we incorporate deep convolutional neural network (CNN), residual learning, and network in network techniques for feature extraction and restoration. In contrast to the current trend of increasing network depth and complexity to boost the imaging performance, we apply a parallel 1×1 CNN to compress the output of the hidden layer and optimize the number of layers and the number of filters for each convolutional layer. The quantitative and qualitative evaluative results demonstrate that our proposed model is accurate, efficient and robust for super-resolution (SR) image restoration from noisy LR input images. In particular, we validate our composite SR networks on three large-scale CT datasets, and obtain promising results as compared to the other state-of-the-art methods.

Automatic Detection of Cone Photoreceptors With Fully Convolutional Networks

PURPOSE

To develop a fully automatic method, based on deep learning algorithms, for determining the locations of cone photoreceptors within adaptive optics scanning laser ophthalmoscope images and evaluate its performance against a dataset of manually segmented images.

METHODS

A fully convolutional network (FCN) based on U-Net architecture was used to generate prediction probability maps and then used a localization algorithm to reduce the prediction map to a collection of points. The proposed method was trained and tested on two publicly available datasets of different imaging modalities, with Dice overlap, false discovery rate, and true positive reported to assess performance.

RESULTS

The proposed method achieves a Dice coefficient of 0.989, true positive rate of 0.987, and false discovery rate of 0.009 on the first confocal dataset; and a Dice coefficient of 0.926, true positive rate of 0.909, and false discovery rate of 0.051 on the second split detector dataset. Results compare favorably with a previously proposed method, but this method provides quicker (25 times faster) evaluation performance.

CONCLUSIONS

The proposed FCN-based method demonstrates that deep learning algorithms can achieve accurate cone localizations, almost comparable to a human expert, while labeling the images.

TRANSLATIONAL RELEVANCE

Manual cone photoreceptor identification is a time-consuming task due to the large number of cones present within a single image; using the proposed FCN-based method could support the image analysis task, drastically reducing the need for manual assessment of the photoreceptor mosaic.

Automatic choroidal segmentation in OCT images using supervised deep learning methods

The analysis of the choroid in the eye is crucial for our understanding of a range of ocular diseases and physiological processes. Optical coherence tomography (OCT) imaging provides the ability to capture highly detailed cross-sectional images of the choroid yet only a very limited number of commercial OCT instruments provide methods for automatic segmentation of choroidal tissue. Manual annotation of the choroidal boundaries is often performed but this is impractical due to the lengthy time taken to analyse large volumes of images. Therefore, there is a pressing need for reliable and accurate methods to automatically segment choroidal tissue boundaries in OCT images. In this work, a variety of patch-based and fully-convolutional deep learning methods are proposed to accurately determine the location of the choroidal boundaries of interest. The effect of network architecture, patch-size and contrast enhancement methods was tested to better understand the optimal architecture and approach to maximize performance. The results are compared with manual boundary segmentation used as a ground-truth, as well as with a standard image analysis technique. Results of total retinal layer segmentation are also presented for comparison purposes. The findings presented here demonstrate the benefit of deep learning methods for segmentation of the chorio-retinal boundary analysis in OCT images.

Development and validation of a deep learning algorithm for distinguishing the nonperfusion area from signal reduction artifacts on OCT angiography

The capillary nonperfusion area (NPA) is a key quantifiable biomarker in the evaluation of diabetic retinopathy (DR) using optical coherence tomography angiography (OCTA). However, signal reduction artifacts caused by vitreous floaters, pupil vignetting, or defocus present significant obstacles to accurate quantification. We have developed a convolutional neural network, MEDnet-V2, to distinguish NPA from signal reduction artifacts in 6×6 mm² OCTA. The network achieves strong specificity and sensitivity for NPA detection across a wide range of DR severity and scan quality.

Joint retina segmentation and classification for early glaucoma diagnosis

We propose a joint segmentation and classification deep model for early glaucoma diagnosis using retina imaging with optical coherence tomography (OCT). Our motivation roots in the observation that ophthalmologists make the clinical decision by analyzing the retinal nerve fiber layer (RNFL) from OCT images. To simulate this process, we propose a novel deep model that joins the retinal layer segmentation and glaucoma classification. Our model consists of three parts. First, the segmentation network simultaneously predicts both six retinal layers and five boundaries between them. Then, we introduce a post processing algorithm to fuse the two results while enforcing the topology correctness. Finally, the classification network takes the RNFL thickness vector as input and outputs the probability of being glaucoma. In the classification network, we propose a carefully designed module to implement the clinical strategy to diagnose glaucoma. We validate our method both in a collected dataset of 1004 circular OCT B-Scans from 234 subjects and in a public dataset of 110 B-Scans from 10 patients with diabetic macular edema. Experimental results demonstrate that our method achieves superior segmentation performance than other state-of-the-art methods both in our collected dataset and in public dataset with severe retina pathology. For glaucoma classification, our model achieves diagnostic accuracy of 81.4% with AUC of 0.864, which clearly outperforms baseline methods.

Parallel deep neural networks for endoscopic OCT image segmentation

We report parallel-trained deep neural networks for automated endoscopic OCT image segmentation feasible even with a limited training data set. These U-Net-based deep neural networks were trained using a modified dice loss function and manual segmentations of ultrahigh-resolution cross-sectional images collected by an 800 nm OCT endoscopic system. The method was tested on in vivo guinea pig esophagus images. Results showed its robust layer segmentation capability with a boundary error of 1.4 µm insensitive to lay topology disorders. To further illustrate its clinical potential, the method was applied to differentiating in vivo OCT esophagus images from an eosinophilic esophagitis (EOE) model and its control group, and the results clearly demonstrated quantitative changes in the top esophageal layers' thickness in the EOE model.

Robust segmentation of retinal layers in optical coherence tomography images based on a multistage active contour model

Optical Coherence Tomography (OCT) constitutes an imaging technique that is increasing its popularity in the ophthalmology field, since it offers a more complete set of information about the main retinal structures. Hence, it offers detailed information about the eye fundus morphology, allowing the identification of many intraretinal pathological signs. For that reason, over the recent years, Computer-Aided Diagnosis (CAD) systems have spread to work with this image modality and analyze its information. A crucial step for the analysis of the retinal tissues implies the identification and delimitation of the different retinal layers. In this context, we present in this work a fully automatic method for the identification of the main retinal layers that delimits the retinal region. Thus, an active contour-based model was completely adapted and optimized to segment these main retinal boundaries. This fully automatic method uses the information of the horizontal placement of these retinal layers and their relative location over the analyzed images to restrict the search space, considering the presence of shadows that are normally generated by pathological or non-pathological artifacts. The validation process was done using the groundtruth of an expert ophthalmologist analyzing healthy as well as unhealthy patients with different degrees of diabetic retinopathy (without macular edema, with macular edema and with lesions in the photoreceptor layers). Quantitative results are in line with the state of the art of this domain, providing accurate segmentations of the retinal layers even when significative pathological alterations are present in the eye fundus. Therefore, the proposed method is robust enough to be used in complex environments, making it feasible for the ophthalmologists in their routine clinical practice.

Artificial intelligence and deep learning in ophthalmology

Artificial intelligence (AI) based on deep learning (DL) has sparked tremendous global interest in recent years. DL has been widely adopted in image recognition, speech recognition and natural language processing, but is only beginning to impact on healthcare. In ophthalmology, DL has been applied to fundus photographs, optical coherence tomography and visual fields, achieving robust classification performance in the detection of diabetic retinopathy and retinopathy of prematurity, the glaucoma-like disc, macular oedema and age-related macular degeneration. DL in ocular imaging may be used in conjunction with telemedicine as a possible solution to screen, diagnose and monitor major eye diseases for patients in primary care and community settings. Nonetheless, there are also potential challenges with DL application in ophthalmology, including clinical and technical challenges, explainability of the algorithm results, medicolegal issues, and physician and patient acceptance of the AI 'black-box' algorithms. DL could potentially revolutionise how ophthalmology is practised in the future. This review provides a summary of the state-of-the-art DL systems described for ophthalmic applications, potential challenges in clinical deployment and the path forward.

Choroidal changes in human myopia: insights from optical coherence tomography imaging

The choroid is a vascular tissue which plays a range of critical roles in the normal physiology of the eye, such as supplying the outer retina with oxygen and nutrients and the regulation of intraocular pressure. There is also substantial evidence, particularly from animal studies, that the choroid plays an important role in the regulation of eye growth and the development of common refractive errors like myopia. In recent years, advances in optical coherence tomography technology have improved our ability to image and measure the choroid in the human eye. Research using this technology over the past decade has dramatically improved our knowledge of the normal choroid, and its potential role in the regulation of eye growth and refractive error development. This review aims to provide an overview of recent work examining the normal human choroid, its changes with myopia and the possible role of the choroid in the mechanism regulating eye growth. Studies have demonstrated that choroidal thinning accompanies the development and progression of myopia, and have established a close link between eye growth and choroidal thickness changes. Dramatic thinning of the choroid is seen with high myopia, and associations are also observed between choroidal thinning and reduced vision, and the development of retinal pathology associated with high myopia. In the short-term, environmental factors known to be associated with myopia development and more rapid eye growth typically lead to a thinning of the choroid, whereas factors linked to a slowing of eye growth are typically associated with short-term choroidal thickening. Collectively, these findings suggest that the choroid is an important biomarker of eye growth in the human eye, and additional research to better understand the human choroid is likely to further our knowledge of the signals and pathways regulating eye growth, myopia development and progression.

Automatic segmentation of OCT retinal boundaries using recurrent neural networks and graph search

The manual segmentation of individual retinal layers within optical coherence tomography (OCT) images is a time-consuming task and is prone to errors. The investigation into automatic segmentation methods that are both efficient and accurate has seen a variety of methods proposed. In particular, recent machine learning approaches have focused on the use of convolutional neural networks (CNNs). Traditionally applied to sequential data, recurrent neural networks (RNNs) have recently demonstrated success in the area of image analysis, primarily due to their usefulness to extract temporal features from sequences of images or volumetric data. However, their potential use in OCT retinal layer segmentation has not previously been reported, and their direct application for extracting spatial features from individual 2D images has been limited. This paper proposes the use of a recurrent neural network trained as a patch-based image classifier (retinal boundary classifier) with a graph search (RNN-GS) to segment seven retinal layer boundaries in OCT images from healthy children and three retinal layer boundaries in OCT images from patients with age-related macular degeneration (AMD). The optimal architecture configuration to maximize classification performance is explored. The results demonstrate that a RNN is a viable alternative to a CNN for image classification tasks in the case where the images exhibit a clear sequential structure. Compared to a CNN, the RNN showed a slightly superior average generalization classification accuracy. Secondly, in terms of segmentation, the RNN-GS performed competitively against a previously proposed CNN based method (CNN-GS) with respect to both accuracy and consistency. These findings apply to both normal and AMD data. Overall, the RNN-GS method yielded superior mean absolute errors in terms of the boundary position with an average error of 0.53 pixels (normal) and 1.17 pixels (AMD). The methodology and results described in this paper may assist the future investigation of techniques within the area of OCT retinal segmentation and highlight the potential of RNN methods for OCT image analysis.