Globally optimal OCT surface segmentation using a constrained IPM optimization

Dense Multi-Scale Graph Convolutional Network for Knee Joint Cartilage Segmentation

In this paper, we propose a dense multi-scale adaptive graph convolutional network (DMA-GCN) method for automatic segmentation of the knee joint cartilage from MR images. Under the multi-atlas setting, the suggested approach exhibits several novelties, as described in the following. First, our models integrate both local-level and global-level learning simultaneously. The local learning task aggregates spatial contextual information from aligned spatial neighborhoods of nodes, at multiple scales, while global learning explores pairwise affinities between nodes, located globally at different positions in the image. We propose two different structures of building models, whereby the local and global convolutional units are combined by following an alternating or a sequential manner. Secondly, based on the previous models, we develop the DMA-GCN network, by utilizing a densely connected architecture with residual skip connections. This is a deeper GCN structure, expanded over different block layers, thus being capable of providing more expressive node feature representations. Third, all units pertaining to the overall network are equipped with their individual adaptive graph learning mechanism, which allows the graph structures to be automatically learned during training. The proposed cartilage segmentation method is evaluated on the entire publicly available Osteoarthritis Initiative (OAI) cohort. To this end, we have devised a thorough experimental setup, with the goal of investigating the effect of several factors of our approach on the classification rates. Furthermore, we present exhaustive comparative results, considering traditional existing methods, six deep learning segmentation methods, and seven graph-based convolution methods, including the currently most representative models from this field. The obtained results demonstrate that the DMA-GCN outperforms all competing methods across all evaluation measures, providing DSC=95.71% and DSC=94.02% for the segmentation of femoral and tibial cartilage, respectively.

Simultaneous alignment and surface regression using hybrid 2D-3D networks for 3D coherent layer segmentation of retinal OCT images with full and sparse annotations

Layer segmentation is important to quantitative analysis of retinal optical coherence tomography (OCT). Recently, deep learning based methods have been developed to automate this task and yield remarkable performance. However, due to the large spatial gap and potential mismatch between the B-scans of an OCT volume, all of them were based on 2D segmentation of individual B-scans, which may lose the continuity and diagnostic information of the retinal layers in 3D space. Besides, most of these methods required dense annotation of the OCT volumes, which is labor-intensive and expertise-demanding. This work presents a novel framework based on hybrid 2D-3D convolutional neural networks (CNNs) to obtain continuous 3D retinal layer surfaces from OCT volumes, which works well with both full and sparse annotations. The 2D features of individual B-scans are extracted by an encoder consisting of 2D convolutions. These 2D features are then used to produce the alignment displacement vectors and layer segmentation by two 3D decoders coupled via a spatial transformer module. Two losses are proposed to utilize the retinal layers' natural property of being smooth for B-scan alignment and layer segmentation, respectively, and are the key to the semi-supervised learning with sparse annotation. The entire framework is trained end-to-end. To the best of our knowledge, this is the first work that attempts 3D retinal layer segmentation in volumetric OCT images based on CNNs. Experiments on a synthetic dataset and three public clinical datasets show that our framework can effectively align the B-scans for potential motion correction, and achieves superior performance to state-of-the-art 2D deep learning methods in terms of both layer segmentation accuracy and cross-B-scan 3D continuity in both fully and semi-supervised settings, thus offering more clinical values than previous works.

Arterial hypertension and retinal layer thickness: the Beijing Eye Study

PURPOSE

To investigate relationships between blood pressure and the thickness of single retinal layers in the macula.

METHODS

Participants of the population-based Beijing Eye Study, free of retinal or optic nerve disease, underwent medical and ophthalmological examinations including optical coherence tomographic examination of the macula. Applying a multiple-surface segmentation solution, we automatically segmented the retina into its various layers.

RESULTS

The study included 2237 participants (mean age 61.8±8.4 years, range 50-93 years). Mean thicknesses of the retinal nerve fibre layer (RNFL), ganglion cell layer (GCL), inner plexiform layer, inner nuclear layer (INL), outer plexiform layer, outer nuclear layer/external limiting membrane, ellipsoid zone, photoreceptor outer segments (POS) and retinal pigment epithelium-Bruch membrane were 31.1±2.3 µm, 39.7±3.5 µm, 38.4±3.3 µm, 34.8±2.0 µm, 28.1±3.0 µm, 79.2±7.3 µm, 22.9±0.6 µm, 19.2±3.3 µm and 20.7±1.4 µm, respectively. In multivariable analysis, higher systolic blood pressure (SBP) and diastolic blood pressure (DBP) were associated with thinner GCL and thicker INL, after adjusting for age, sex and axial length (all p<0.0056). Higher SBP was additionally associated with thinner POS and higher DBP with thinner RNFL. For an elevation of SBP/DBP by 10 mm Hg, the RNFL, GCL, INL and POS changed by 2.0, 3.0, 1.5 and 2.0 µm, respectively.

CONCLUSIONS

Thickness of RNFL, GCL and POS was inversely and INL thickness was positively associated with higher blood pressure, while the thickness of the other retinal layers was not significantly correlated with blood pressure. The findings may be helpful for refinement of the morphometric detection of retinal diseases.

Deep learning network with differentiable dynamic programming for retina OCT surface segmentation

Multiple-surface segmentation in optical coherence tomography (OCT) images is a challenging problem, further complicated by the frequent presence of weak image boundaries. Recently, many deep learning-based methods have been developed for this task and yield remarkable performance. Unfortunately, due to the scarcity of training data in medical imaging, it is challenging for deep learning networks to learn the global structure of the target surfaces, including surface smoothness. To bridge this gap, this study proposes to seamlessly unify a U-Net for feature learning with a constrained differentiable dynamic programming module to achieve end-to-end learning for retina OCT surface segmentation to explicitly enforce surface smoothness. It effectively utilizes the feedback from the downstream model optimization module to guide feature learning, yielding better enforcement of global structures of the target surfaces. Experiments on Duke AMD (age-related macular degeneration) and JHU MS (multiple sclerosis) OCT data sets for retinal layer segmentation demonstrated that the proposed method was able to achieve subvoxel accuracy on both datasets, with the mean absolute surface distance (MASD) errors of 1.88 ± 1.96μm and 2.75 ± 0.94μm, respectively, over all the segmented surfaces.

Macular outer nuclear layer, ellipsoid zone and outer photoreceptor segment band thickness, axial length and other determinants

The study aims to assess the thickness of the retinal outer nuclear layer (ONL), ellipsoid zone (EZ) and photoreceptor outer segment (POS) band in various macular regions and its associations with axial length and other parameters. Participants of the Beijing Eye Study 2011 underwent a series of examinations including spectral-domain optical coherence tomography of the macula. The current study included 2213 participants without retinal or optic nerve diseases (age: 61.7 ± 8.4 years; range 50-93 years); axial length: 23.15 ± 0.95 mm; range 18.96-29.15 mm). The ONL (fovea: 98.9 ± 8.8 µm), EZ (fovea: 24.1 ± 0.5 µm) and POS band (fovea: 24.3 ± 3.5 µm) were the thickest (P < 0.001) in the fovea (defined as the thinnest central point), followed by the temporal inner, nasal inner, inferior inner, superior inner, inferior outer, temporal outer, nasal outer, and superior outer region. In multivariable analysis, a thicker retinal ONL was associated (correlation coefficient r: 0.40) with shorter axial length (beta: - 0.14; P < 0.001) and shorter disc-fovea distance (beta: - 0.10; P = 0.001), after adjusting for younger age (beta: - 0.26; P < 0.001), male sex (beta: 0.24; P < 0.001), lower serum cholesterol concentration (beta: - 0.05; P = 0.04), and thicker subfoveal choroidal thickness (beta: 0.08; P < 0.001). The POS thickness increased with shorter axial length (beta: - 0.06; P < 0.001) and shorter optic disc-fovea distance (beta: - 0.05; P = 0.03), after adjusting for younger age (beta: - 0.34; P < 0.001), male sex (beta: 0.15; P < 0.001), and thicker subfoveal choroidal thickness (beta: 0.24; P < 0.001). As a conclusion, the photoreceptor ONL, EZ and POS band vary in thickness between different macular regions and differ in their correlations with axial length, disc-fovea distance, age, sex, and subfoveal choroidal thickness. The ONL thickness decrease with longer axial length and longer disc-fovea distance may point to an axial elongation-associated retinal stretching in the macula.

KCB-Net: A 3D knee cartilage and bone segmentation network via sparse annotation

Knee cartilage and bone segmentation is critical for physicians to analyze and diagnose articular damage and knee osteoarthritis (OA). Deep learning (DL) methods for medical image segmentation have largely outperformed traditional methods, but they often need large amounts of annotated data for model training, which is very costly and time-consuming for medical experts, especially on 3D images. In this paper, we report a new knee cartilage and bone segmentation framework, KCB-Net, for 3D MR images based on sparse annotation. KCB-Net selects a small subset of slices from 3D images for annotation, and seeks to bridge the performance gap between sparse annotation and full annotation. Specifically, it first identifies a subset of the most effective and representative slices with an unsupervised scheme; it then trains an ensemble model using the annotated slices; next, it self-trains the model using 3D images containing pseudo-labels generated by the ensemble method and improved by a bi-directional hierarchical earth mover's distance (bi-HEMD) algorithm; finally, it fine-tunes the segmentation results using the primal-dual Internal Point Method (IPM). Experiments on four 3D MR knee joint datasets (the SKI10 dataset, OAI ZIB dataset, Iowa dataset, and iMorphics dataset) show that our new framework outperforms state-of-the-art methods on full annotation, and yields high quality results for small annotation ratios even as low as 10%.

DPSF: a Novel Dual-Parametric Sigmoid Function for Optical Coherence Tomography Image Enhancement

Speckle noise reduces the image contrast significantly making the highly scattering structures boundaries difficult to distinguish. This has limited the usage of optical coherence tomography (OCT) images in clinical routine and hindered its potential by depriving clinicians from assessing useful information that are needed in disease monitoring, treatment, progression, and decision making. To overcome this limitation, we propose a fast and robust OCT image enhancement framework using non-linear statistical parametric technique. In the proposed framework, we utilize prior statistical information to model the image to follow Gaussian distribution. After which, a newly designed dual-parametric sigmoid function (DPSF) is utilized to control the dynamic range and contrast level of the image. To balance the intensity range and contrast level, both linear and non-linear normalization operations are performed, then followed by a mapping operation to obtain the enhanced image. Experimentation results on the three OCT vendors show that the proposed method obtained high values in EME, PSNR, SSIM, ρ, and low value in MSE of 36.72, 38.87, 0.87, 0.98, and 25.12 for Cirrus; 40.77, 41.84, 0.89, 0.98, and 22.15 for Spectralis; and 30.81, 32.10, 0.81, 0.96, and 28.55 for Topcon OCT devices, respectively. The proposed DPSF framework performs better than the state-of-the-art methods and improves the interpretability and perception of the OCT images, which can provide clinicians and computer vision program with good quantitative and qualitative information.