Optimal surface segmentation in volumetric images--a graph-theoretic approach

Automated segmentation of choroidal layers from 3-dimensional macular optical coherence tomography scans

BACKGROUND

Changes in choroidal thickness are associated with various ocular diseases, and the choroid can be imaged using spectral-domain optical coherence tomography (SD-OCT) and enhanced depth imaging OCT (EDI-OCT).

NEW METHOD

Eighty macular SD-OCT volumes from 80 patients were obtained using the Zeiss Cirrus machine. Eleven additional control subjects had two Cirrus scans done in one visit along with enhanced depth imaging (EDI-OCT) using the Heidelberg Spectralis machine. To automatically segment choroidal layers from the OCT volumes, our graph-theoretic approach was utilized. The segmentation results were compared with reference standards from two independent graders, and the accuracy of automated segmentation was calculated using unsigned/signed border positioning/thickness errors and Dice similarity coefficient (DSC). The repeatability and reproducibility of our choroidal thicknesses were determined by intraclass correlation coefficient (ICC), coefficient of variation (CV), and repeatability coefficient (RC).

RESULTS

The mean unsigned/signed border positioning errors for the choroidal inner and outer surfaces are 3.39 ± 1.26 µm (mean ± standard deviation)/- 1.52 ± 1.63 µm and 16.09 ± 6.21 µm/4.73 ± 9.53 µm, respectively. The mean unsigned/signed choroidal thickness errors are 16.54 ± 6.47 µm/6.25 ± 9.91 µm, and the mean DSC is 0.949 ± 0.025. The ICC (95% confidence interval), CV, RC values are 0.991 (0.977-0.997), 2.48%, 14.25 µm for the repeatability and 0.991 (0.977-0.997), 2.49%, 14.30 µm for the reproducibility studies, respectively.

COMPARISON WITH EXISTING METHOD(S)

The proposed method outperformed our previous method using choroidal vessel segmentation and inter-grader variability.

CONCLUSIONS

This automated segmentation method can reliably measure choroidal thickness using different OCT platforms.

CNN-based CP-OCT sensor integrated with a subretinal injector for retinal boundary tracking and injection guidance

SIGNIFICANCE

Subretinal injection is an effective way of delivering transplant genes and cells to treat many degenerative retinal diseases. However, the technique requires high-dexterity and microscale precision of experienced surgeons, who have to overcome the physiological hand tremor and limited visualization of the subretinal space.

AIM

To automatically guide the axial motion of microsurgical tools (i.e., a subretinal injector) with microscale precision in real time using a fiber-optic common-path swept-source optical coherence tomography distal sensor.

APPROACH

We propose, implement, and study real-time retinal boundary tracking of A-scan optical coherence tomography (OCT) images using a convolutional neural network (CNN) for automatic depth targeting of a selected retinal boundary for accurate subretinal injection guidance. A simplified 1D U-net is used for the retinal layer segmentation on A-scan OCT images. A Kalman filter, combining retinal boundary position measurement by CNN and velocity measurement by cross correlation between consecutive A-scan images, is applied to optimally estimate the retinal boundary position. Unwanted axial motions of the surgical tools are compensated by a piezoelectric linear motor based on the retinal boundary tracking.

RESULTS

CNN-based segmentation on A-scan OCT images achieves the mean unsigned error (MUE) of ∼3  pixels (8.1  μm) using an ex vivo bovine retina model. GPU parallel computing allows real-time inference (∼2  ms) and thus real-time retinal boundary tracking. Involuntary tremors, which include low-frequency draft in hundreds of micrometers and physiological tremors in tens of micrometers, are compensated effectively. The standard deviations of photoreceptor (PR) and choroid (CH) boundary positions get as low as 10.8  μm when the depth targeting is activated.

CONCLUSIONS

A CNN-based common-path OCT distal sensor successfully tracks retinal boundaries, especially the PR/CH boundary for subretinal injection, and automatically guides the tooltip's axial position in real time. The microscale depth targeting accuracy of our system shows its promising possibility for clinical application.

Automatic Segmentation in Multiple OCT Layers For Stargardt Disease Characterization Via Deep Learning

Purpose

This study sought to perform automated segmentation of 11 retinal layers and Stargardt-associated features on spectral-domain optical coherence tomography (SD-OCT) images and to analyze differences between normal eyes and eyes diagnosed with Stargardt disease.

Methods

Automated segmentation was accomplished through application of the deep learning-shortest path (DL-SP) framework, a shortest path segmentation approach that is enhanced by a deep learning fully convolutional neural network. To compare normal eyes and eyes diagnosed with Stargardt disease, various retinal layer thickness and intensity feature maps associated with the outer retinal layers were generated.

Results

The automated DL-SP approach achieved a mean difference within a subpixel accuracy range for all layers when compared to manually traced layers by expert graders. The algorithm achieved mean and absolute mean differences in border positions for Stargardt features of -0.11 ± 4.17 pixels and 1.92 ± 3.71 pixels, respectively. In several of the feature maps generated, the characteristic Stargardt features of flecks and atrophic-appearing lesions were readily visualized.

Conclusions

To the best of our knowledge, this is the first automated algorithm for 11 retinal layer segmentation on OCT in eyes with Stargardt disease, and, furthermore, the feature differences found between eyes diagnosed with Stargardt disease and normal eyes may inform new insights and the better understanding of retinal characteristic morphologic changes caused by Stargardt disease.

Translational Relevance

The automated algorithm's performance and the feature differences found using the algorithm's segmentation support the future applications of SD-OCT for the quantitative monitoring of Stargardt disease.

A novel quantitative analysis method for idiopathic epiretinal membrane

PURPOSE

To introduce a novel method to quantitively analyse in three dimensions traction forces in a vast area of the ocular posterior pole.

METHODS

Retrospective analysis of 14 eyes who underwent peeling surgery for idiopathic, symptomatic and progressive epiretinal membrane. The technique measures the shift in position of vascular crossings after surgery from a fixed point, which is the retinal pigmented epithelium. This shift is defined as the relaxation index (RI) and represents a measure of the postoperative movement of the retina due to released traction after surgery.

RESULTS

Best-corrected visual acuity was significantly better than baseline at all follow ups while the RI had its maximum value at baseline. Moreover, we found a significant correlation between best-corrected visual acuity at 6 months and RI at baseline.

CONCLUSION

While all previous published methods focused on bi-dimensional changes observed in a small region, this study introduces a three-dimensional assessment of tractional forces. Future integration of RI into built-in processing software will allow systematic three-dimensional measurement of intraretinal traction.

Structured layer surface segmentation for retina OCT using fully convolutional regression networks

Optical coherence tomography (OCT) is a noninvasive imaging modality with micrometer resolution which has been widely used for scanning the retina. Retinal layers are important biomarkers for many diseases. Accurate automated algorithms for segmenting smooth continuous layer surfaces with correct hierarchy (topology) are important for automated retinal thickness and surface shape analysis. State-of-the-art methods typically use a two step process. Firstly, a trained classifier is used to label each pixel into either background and layers or boundaries and non-boundaries. Secondly, the desired smooth surfaces with the correct topology are extracted by graph methods (e.g., graph cut). Data driven methods like deep networks have shown great ability for the pixel classification step, but to date have not been able to extract structured smooth continuous surfaces with topological constraints in the second step. In this paper, we combine these two steps into a unified deep learning framework by directly modeling the distribution of the surface positions. Smooth, continuous, and topologically correct surfaces are obtained in a single feed forward operation. The proposed method was evaluated on two publicly available data sets of healthy controls and subjects with either multiple sclerosis or diabetic macular edema, and is shown to achieve state-of-the art performance with sub-pixel accuracy.

Accurately motion-corrected Lissajous OCT with multi-type image registration

Passive motion correction methods for optical coherence tomography (OCT) use image registration to estimate eye movements. To improve motion correction, a multi-image cross-correlation that employs spatial features in different image types is introduced. Lateral motion correction using en face OCT and OCT-A projections on Lissajous-scanned OCT data is applied. Motion correction using OCT-A projection of whole depth and OCT amplitude, OCT logarithmic intensity, and OCT maximum intensity projections were evaluated in retinal imaging with 76 patients. The proposed method was compared with motion correction using OCT-A projection of whole depth. The comparison shows improvements in the image quality of motion-corrected superficial OCT-A images and image registration.

Effects of High-Intensity Exercise Training on Adipose Tissue Mass, Glucose Uptake and Protein Content in Pre- and Post-menopausal Women

The menopausal transition is accompanied by changes in adipose tissue storage, leading to an android body composition associated with increased risk of type 2 diabetes and cardiovascular disease in post-menopausal women. Estrogens probably affect local adipose tissue depots differently. We investigated how menopausal status and exercise training influence adipose tissue mass, adipose tissue insulin sensitivity and adipose tissue proteins associated with lipogenesis/lipolysis and mitochondrial function. Healthy, normal-weight pre- (n = 21) and post-menopausal (n = 20) women participated in high-intensity exercise training three times per week for 12 weeks. Adipose tissue distribution was determined by dual-energy x-ray absorptiometry and magnetic resonance imaging. Adipose tissue glucose uptake was assessed by positron emission tomography/computed tomography (PET/CT) by the glucose analog [18F]fluorodeoxyglucose ([18F]FDG) during continuous insulin infusion (40 mU·m⁻²·min⁻¹). Protein content associated with insulin signaling, lipogenesis/lipolysis, and mitochondrial function were determined by western blotting in abdominal and femoral white adipose tissue biopsies. The mean age difference between the pre- and the post-menopausal women was 4.5 years. Exercise training reduced subcutaneous (~4%) and visceral (~6%) adipose tissue masses similarly in pre- and post-menopausal women. Insulin-stimulated glucose uptake, assessed by [18F]FDG-uptake during PET/CT, was similar in pre- and post-menopausal women in abdominal, gluteal, and femoral adipose tissue depots, despite skeletal muscle insulin resistance in post- compared to pre-menopausal women in the same cohort. Insulin-stimulated glucose uptake in adipose tissue depots was not changed after 3 months of high-intensity exercise training, but insulin sensitivity was higher in visceral compared to subcutaneous adipose tissue depots (~139%). Post-menopausal women exhibited increased hexokinase and adipose triglyceride lipase content in subcutaneous abdominal adipose tissue. Physical activity in the early post-menopausal years reduces abdominal obesity, but insulin sensitivity of adipose tissue seems unaffected by both menopausal status and physical activity.

End-to-End Deep Learning Model for Predicting Treatment Requirements in Neovascular AMD From Longitudinal Retinal OCT Imaging

Neovascular age-related macular degeneration (nAMD) is nowadays successfully treated with anti-VEGF substances, but inter-individual treatment requirements are vastly heterogeneous and currently poorly plannable resulting in suboptimal treatment frequency. Optical coherence tomography (OCT) with its 3D high-resolution imaging serves as a companion diagnostic to anti-VEGF therapy. This creates a need for building predictive models using automated image analysis of OCT scans acquired during the treatment initiation phase. We propose such a model based on deep learning (DL) architecture, comprised of a densely connected neural network (DenseNet) and a recurrent neural network (RNN), trainable end-to-end. The method starts by sampling several 2D-images from an OCT volume to obtain a lower-dimensional OCT representation. At the core of the predictive model, the DenseNet learns useful retinal spatial features while the RNN integrates information from different time points. The introduced model was evaluated on the prediction of anti-VEGF treatment requirements in nAMD patients treated under a pro-re-nata (PRN) regimen. The DL model was trained on 281 patients and evaluated on a hold-out test set of 69 patient. The predictive model achieved a concordance index of 0.7 in regressing the number of received treatments, while in a classification task it obtained an 0.85 (0.81) AUC in detecting the patients with low (high) treatment requirements. The proposed model outperformed previous machine learning strategies that relied on a set of spatio-temporal image features, showing that the proposed DL architecture successfully learned to extract the relevant spatio-temporal patterns directly from raw longitudinal OCT images.

3D Retinal Vessel Density Mapping With OCT-Angiography

Optical Coherence Tomography Angiography (OCTA) is a novel, non-invasive imaging modality of retinal capillaries at micron resolution. Recent studies have correlated macular OCTA vascular measures with retinal disease severity and supported their use as a diagnostic tool. However, these measurements mostly rely on a few summary statistics in retinal layers or regions of interest in the two-dimensional (2D) en face projection images. To enable 3D and localized comparisons of retinal vasculature between longitudinal scans and across populations, we develop a novel approach for mapping retinal vessel density from OCTA images. We first obtain a high-quality 3D representation of OCTA-based vessel networks via curvelet-based denoising and optimally oriented flux (OOF). Then, an effective 3D retinal vessel density mapping method is proposed. In this framework, a vessel density image (VDI) is constructed by diffusing the vessel mask derived from OOF-based analysis to the entire image volume. Subsequently, we utilize a non-linear, 3D OCT image registration method to provide localized comparisons of retinal vasculature across subjects. In our experimental results, we demonstrate an application of our method for longitudinal qualitative analysis of two pathological subjects with edema during the course of clinical care. Additionally, we quantitatively validate our method on synthetic data with simulated capillary dropout, a dataset obtained from a normal control (NC) population divided into two age groups and a dataset obtained from patients with diabetic retinopathy (DR). Our results show that we can successfully detect localized vascular changes caused by simulated capillary loss, normal aging, and DR pathology even in presence of edema. These results demonstrate the potential of the proposed framework in localized detection of microvascular changes and monitoring retinal disease progression.

Retinal layer thicknesses retrieved with different segmentation algorithms from optical coherence tomography scans acquired under different signal-to-noise ratio conditions

Glaucomatous damage can be quantified by measuring the thickness of different retinal layers. However, poor image quality may hamper the accuracy of the layer thickness measurement. We determined the effect of poor image quality (low signal-to-noise ratio) on the different layer thicknesses and compared different segmentation algorithms regarding their robustness against this degrading effect. For this purpose, we performed OCT measurements in the macular area of healthy subjects and degraded the image quality by employing neutral density filters. We also analysed OCT scans from glaucoma patients with different disease severity. The algorithms used were: The Canon HS-100's built-in algorithm, DOCTRAP, IOWA, and FWHM, an approach we developed. We showed that the four algorithms used were all susceptible to noise at a varying degree, depending on the retinal layer assessed, and the results between different algorithms were not interchangeable. The algorithms also differed in their ability to differentiate between young healthy eyes and older glaucoma eyes and failed to accurately separate different glaucoma stages from each other.

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.

Increased contrast enhancement of the parent vessel of unruptured intracranial aneurysms in 7T MR imaging

BACKGROUND

Inflammation of the arterial wall may lead to aneurysm formation. The presence of aneurysm enhancement on high-resolution vessel wall imaging (HR-VWI) is a marker of wall inflammation and instability. We aim to determine if there is any association between increased contrast enhancement in the aneurysmal wall and its parent artery.

METHODS

Patients with unruptured intracranial aneurysms (UIAs) prospectively underwent 7T HR-VWI. Regions of interest were selected manually and with a semi-automated protocol based on gradient algorithms of intensity patterns. Mean signal intensities in pre- and post-contrast T1-weighted sequences were adjusted to the enhancement of the pituitary stalk and then subtracted to objectively determine: circumferential aneurysmal wall enhancement (CAWE); parent vessel enhancement (PVE); and reference vessel enhancement (RVE). PVE was assessed over regions located 3- and 5 mm from the aneurysm's neck. RVE was assessed in arteries located in a different vascular territory.

RESULTS

Twenty-five UIAs were analyzed. There was a significant moderate correlation between CAWE and 5 mm PVE (Pearson R=0.52, P=0.008), whereas no correlation was found between CAWE and RVE (Pearson R=0.20, P=0.33). A stronger correlation was found between CAWE and 3 mm PVE (Pearson R=0.78, P<0.001). Intra-class correlation analysis demonstrated good reliability between measurements obtained using semi-automated and manual segmentation (ICC coefficient=0.790, 95% CI 0.58 to 0.90).

CONCLUSION

Parent arteries exhibit higher contrast enhancement in regions closer to the aneurysm's neck, especially in aneurysms≥7 mm. A localized inflammatory/vasculopathic process in the wall of the parent artery may lead to aneurysm formation and growth.

Diagnostic Ability of Individual Macular Layers by Spectral-Domain OCT in Different Stages of Glaucoma

PURPOSE

To compare the diagnostic ability of macular intraretinal layer thickness with circumpapillary retinal nerve fiber layer (cpRNFL) thickness, either when used individually or in combination with cpRNFL for detecting early, moderate, and advanced glaucoma.

DESIGN

Cross-sectional study.

PARTICIPANTS

A total of 423 glaucoma participants and 423 age- and gender-matched normal participants.

METHODS

Participants underwent Cirrus spectral-domain OCT (SD-OCT) imaging (Carl Zeiss Meditec, Dublin, CA) using the optic disc and macular scanning protocols. Iowa Reference Algorithms (version 3.8.0) were used for intraretinal layer segmentation, and mean thickness of intraretinal layers was rescaled with magnification correction using axial length value. Thickness measurements of each layer/sector and their corresponding areas under the receiver operating characteristic curve (AUCs) were obtained. Glaucoma eyes were subdivided based on of their visual field severity (early, n = 234; moderate, n = 107; advanced, n = 82).

MAIN OUTCOME MEASURES

Intraretinal layers.

RESULTS

Some 67% of participants were male, their average ± standard deviation age was 65±9 years. Circumpapillary retinal nerve fiber layer, macular ganglion cell layer (mGCL), and macular inner plexiform layer (mIPL) were significantly thinner in the glaucoma groups (P < 0.0005). The 2 best parameters for detecting normal eyes from early glaucoma was cpRNFL (AUC = 0.861) and mGCL (AUC = 0.842), from moderate glaucoma was mGCL combined with inner plexiform layer (IPL) (AUC = 0.915) and cpRNFL (AUC = 0 .914), and from advanced glaucoma was mGCL-IPL (AUC = 0.984) and cpRNFL (AUC = 0.977). There was no statistical significance between AUCs for the macular parameter and cpRNFL thickness measurement at any of the severities (P > 0.05). Combining macular and cpRNFL parameters improved the diagnostic performance for early glaucoma (AUC = 0.908; P = 0.002) and moderate glaucoma (AUC = 0.944; P = 0.031) but not for advanced glaucoma (AUC = 0.991; P > 0.05).

CONCLUSIONS

Single-layer mGCL thickness is comparable to the traditional cpRNFL thickness for the diagnosis of early/moderate glaucoma, whereas cpRNFL thickness remains the most efficient for advanced glaucoma. Combining macular measurements (GCL and GCL-IPL) and cpRNFL improved the discrimination of early/moderate glaucoma but not of advanced glaucoma. For the diagnosis of early glaucoma, both macular and optic disc scans should be used.

The Retinal Inner Plexiform Synaptic Layer Mirrors Grey Matter Thickness of Primary Visual Cortex with Increased Amyloid β Load in Early Alzheimer's Disease

The retina may serve as putative window into neuropathology of synaptic loss in Alzheimer's disease (AD). Here, we investigated synapse-rich layers versus layers composed by nuclei/cell bodies in an early stage of AD. In addition, we examined the associations between retinal changes and molecular and structural markers of cortical damage. We recruited 20 AD patients and 17 healthy controls (HC). Combining optical coherence tomography (OCT), magnetic resonance (MR), and positron emission tomography (PET) imaging, we measured retinal and primary visual cortex (V1) thicknesses, along with V1 amyloid β (Aβ) retention ([11C]-PiB PET tracer) and neuroinflammation ([11C]-PK11195 PET tracer). We found that V1 showed increased amyloid-binding potential, in the absence of neuroinflammation. Although thickness changes were still absent, we identified a positive association between the synapse-rich inner plexiform layer (IPL) and V1 in AD. This retinocortical interplay might reflect changes in synaptic function resulting from Aβ deposition, contributing to early visual loss.

A Reflectivity Measure to Quantify Bruch's Membrane Calcification in Patients with Pseudoxanthoma Elasticum Using Optical Coherence Tomography

Purpose

Progressive calcification of Bruch's membrane (BM) causes considerable visual morbidity in patients with pseudoxanthoma elasticum (PXE). Since calcification is hyperreflective on optical coherence tomography (OCT), our aim was to measure BM calcification with OCT imaging.

Methods

Case-control study with 45 patients with PXE under 40 years (range, 11-39) and 25 controls (range, 14-39). Spectralis HRA-OCT imaging consisted of seven macular B-scans with 250-µm spacing. Retinal segmentation was performed with the IOWA Reference Algorithms. MATLAB was used to extract and average z-axis reflectivity profiles. Layer reflectivities were normalized to the ganglion cell and inner plexiform layers. Both median and peak layer reflectivities were compared between patients with PXE and controls. The discriminative value of the retinal pigment epithelium (RPE)-BM peak reflectivity was analyzed using receiver operating characteristic analysis.

Results

The reflectivity profile of patients with PXE differed from controls in the outer retinal layers. The normalized median RPE-BM reflectivity was 41.1 (interquartile range [IQR], 26.3-51.9) in patients with PXE, compared with 22.5 (IQR, 19.3-29.5) in controls (P = 2.09 × 10^-3). The normalized RPE-BM peak reflectivity was higher in patients with PXE (67.5; IQR, 42.1-84.2) than in controls (32.7; IQR, 25.7-38.9; P = 2.43 × 10^-5) and had a high discriminative value with an area under the curve of 0.85 (95% confidence interval, 0.76-0.95). In patients with PXE under 40 years, increasing age did not have a statistically significant effect on the RPE-BM peak reflectivity (patients under 20 years: 44.2 [IQR, 40.5-74.6]; 20-30 years: 66.0 [IQR, 45.1-83.8]; 30-40 years: 70.8 [IQR, 49.0-88.0], P = 0.47).

Conclusions

BM calcification can be measured as increased RPE-BM reflectivity in young patients with PXE and has a high discriminative value.

Translational Relevance

In patients with PXE, the OCT reflectivity of Bruch's membrane may be the first biomarker for Bruch's membrane calcification and a valuable ophthalmologic endpoint in clinical trials.

Longitudinal Analysis of Bruch Membrane Opening Morphometry in Myopic Glaucoma

PRéCIS:: The Bruch membrane opening (BMO) was posteriorly bowed and the degree of nonplanarity increased in stable and progressive glaucoma subjects. BMO became more posterior relative to the Bruch membrane (BM) in control and both stable and progressive glaucoma subjects.

PURPOSE

To investigate longitudinal changes in morphologic characteristics of the BMO in control and glaucomatous subjects.

MATERIALS AND METHODS

A total of 53 myopic eyes (17 control, 6 suspect, 20 stable glaucoma, and 10 progressing glaucoma) were followed for an average of 4.2±1.4 years and imaged at the baseline and 2 follow-up appointments using a 1060 nm swept-source optical coherence tomography system. BM and BMO were segmented, and 4 morphometric BMO parameters (area, ellipse ratio, nonplanarity, and depth) were measured.

RESULTS

There were no significant changes in BMO area or ellipse ratio for all groups. BMO nonplanarity was shown to increase in the glaucoma groups. BMO depth relative to BM increased in all groups except the suspects (control: 8.1 µm/y, P=0.0001; stable glaucoma: 3.5 µm/y, P=0.0001; progressing glaucoma: 14.0 µm/y, P=0.0026). In linear mixed-model analysis, axial length was positively associated with BMO area in all groups except for progressing glaucoma, and with BMO nonplanarity in stable glaucoma. It was not a significant factor to the slopes of the BMO parameters in the ANCOVA analysis of slopes.

CONCLUSIONS

Longitudinally, BMO increased in nonplanarity in the glaucoma eyes, and its axial position relative to BM became more posterior in both control and glaucoma eyes.

Automated Retinal Layer Segmentation Using Graph-based Algorithm Incorporating Deep-learning-derived Information

Regular drusen, an accumulation of material below the retinal pigment epithelium (RPE), have long been established as a hallmark early feature of nonneovascular age-related macular degeneration (AMD). Advances in imaging have expanded the phenotype of AMD to include another extracellular deposit, reticular pseudodrusen (RPD) (also termed subretinal drusenoid deposits, SDD), which are located above the RPE. We developed an approach to automatically segment retinal layers associated with regular drusen and RPD in spectral domain (SD) optical coherence tomography (OCT) images. More specifically, a shortest-path algorithm enhanced with probability maps generated through a fully convolutional neural network was used to segment drusen and RPD, as well as 11 retinal layers in SD-OCT volumes. This algorithm achieves a mean difference that is within the subpixel accuracy range drusen and RPD, alongside the other 11 retinal layers, highlighting the high robustness of this algorithm for this dataset. To the best of our knowledge, this is the first report of a validated algorithm for the automated segmentation of the retinal layers including early AMD features of RPD and regular drusen separately on SD-OCT images.

Comparison of the Iowa Reference Algorithm to the Heidelberg Spectralis optical coherence tomography segmentation algorithm

For spectral-domain optical coherence tomography (SD-OCT) studies of neurodegeneration, it is important to understand how segmentation algorithms differ in retinal layer thickness measurements, segmentation error locations and the impact of manual correction. Using macular SD-OCT images of frontotemporal degeneration patients and controls, we compare the individual and aggregate retinal layer thickness measurements provided by two commonly used algorithms, the Iowa Reference Algorithm and Heidelberg Spectralis, with manual correction of significant segmentation errors. We demonstrate small differences of most retinal layer thickness measurements between these algorithms. Outer sectors of the Early Treatment Diabetic Retinopathy Study grid require a greater percent of eyes to be corrected than inner sectors of the retinal nerve fiber layer (RNFL). Manual corrections affect thickness measurements mildly, resulting in at most a 5% change in RNFL thickness. Our findings can inform researchers how to best use different segmentation algorithms when comparing retinal layer thicknesses.

Correction propagation for user-assisted optical coherence tomography segmentation: general framework and application to Bruch's membrane segmentation

Optical coherence tomography (OCT) is a commonly used ophthalmic imaging modality. While OCT has traditionally been viewed cross-sectionally (i.e., as a sequence of B-scans), higher A-scan rates have increased interest in en face OCT visualization and analysis. The recent clinical introduction of OCT angiography (OCTA) has further spurred this interest, with chorioretinal OCTA being predominantly displayed via en face projections. Although en face visualization and quantitation are natural for many retinal features (e.g., drusen and vasculature), it requires segmentation. Because manual segmentation of volumetric OCT data is prohibitively laborious in many settings, there has been significant research and commercial interest in developing automatic segmentation algorithms. While these algorithms have achieved impressive results, the variability of image qualities and the variety of ocular pathologies cause even the most robust automatic segmentation algorithms to err. In this study, we develop a user-assisted segmentation approach, complementary to fully-automatic methods, wherein correction propagation is used to reduce the burden of manually correcting automatic segmentations. The approach is evaluated for Bruch's membrane segmentation in eyes with advanced age-related macular degeneration.

Generative-based airway and vessel morphology quantification on chest CT images

Accurately and precisely characterizing the morphology of small pulmonary structures from Computed Tomography (CT) images, such as airways and vessels, is becoming of great importance for diagnosis of pulmonary diseases. The smaller conducting airways are the major site of increased airflow resistance in chronic obstructive pulmonary disease (COPD), while accurately sizing vessels can help identify arterial and venous changes in lung regions that may determine future disorders. However, traditional methods are often limited due to image resolution and artifacts. We propose a Convolutional Neural Regressor (CNR) that provides cross-sectional measurement of airway lumen, airway wall thickness, and vessel radius. CNR is trained with data created by a generative model of synthetic structures which is used in combination with Simulated and Unsupervised Generative Adversarial Network (SimGAN) to create simulated and refined airways and vessels with known ground-truth. For validation, we first use synthetically generated airways and vessels produced by the proposed generative model to compute the relative error and directly evaluate the accuracy of CNR in comparison with traditional methods. Then, in-vivo validation is performed by analyzing the association between the percentage of the predicted forced expiratory volume in one second (FEV1%) and the value of the Pi10 parameter, two well-known measures of lung function and airway disease, for airways. For vessels, we assess the correlation between our estimate of the small-vessel blood volume and the lungs' diffusing capacity for carbon monoxide (DLCO). The results demonstrate that Convolutional Neural Networks (CNNs) provide a promising direction for accurately measuring vessels and airways on chest CT images with physiological correlates.

Adaptive-Guided-Coupling-Probability Level Set for Retinal Layer Segmentation

Quantitative assessment of retinal layer thickness in spectral domain-optical coherence tomography (SD-OCT) images is vital for clinicians to determine the degree of ophthalmic lesions. However, due to the complex retinal tissues, high-level speckle noises and low intensity constraint, how to accurately recognize the retinal layer structure still remains a challenge. To overcome this problem, this paper proposes an adaptive-guided-coupling-probability level set method for retinal layer segmentation in SD-OCT images. Specifically, based on Bayes's theorem, each voxel probability representation is composed of two probability terms in our method. The first term is constructed as neighborhood Gaussian fitting distribution to characterize intensity information for each intra-retinal layer. The second one is boundary probability map generated by combining anatomical priors and adaptive thickness information to ensure surfaces evolve within a proper range. Then, the voxel probability representation is introduced into the proposed segmentation framework based on coupling probability level set to detect layer boundaries. A total of 1792 retinal B-scan images from 4 SD-OCT cubes in healthy eyes, 5 cubes in abnormal eyes with central serous chorioretinaopathy and 5 SD-OCT cubes in abnormal eyes with age-related macular disease are used to evaluate the proposed method. The experiment demonstrates that the segmentation results obtained by the proposed method have a good consistency with ground truth, and the proposed method outperforms six methods in the layer segmentation of uneven retinal SD-OCT images.

The Detection of Retina Microvascular Density in Subclinical Aquaporin-4 Antibody Seropositive Neuromyelitis Optica Spectrum Disorders

Purpose: To use optical coherence tomography (OCT) and OCT angiography (OCT-A) to measure changes in the retinal structure and microvasculature of patients with aquaporin-4 antibody-positive, neuromyelitis optica spectrum disorder (NMOSD) with a history of optic neuritis (NMOSD+ON) and those without it (NMOSD–ON).

Methods: A total of 27 aquaporin-4 antibody-positive NMOSD patients and 31 age- and gender-matched healthy control (HC) participants were included. In 27 NMOSD patients, 19 of them had a history of optic neuritis (ON) and 8 of them had no history of ON. Peripapillary retinal nerve fiber layer (pRNFL) thickness and macular ganglion cell and inner plexiform layer (GCIPL) thickness were measured by OCT. Radial peripapillary capillary density (RPCD) and macular superficial vessel density (MSVD) were measured by OCT-A. Comparisons of retinal structural and microvascular parameters between the cohorts were performed using generalized estimating equation (GEE) models. Diagnostic accuracy was evaluated by the area under the receiver operating characteristics curve (AROC).

Results: In NMOSD+ON eyes, the GCIPL and pRNFL thicknesses, 48.6 ± 7.1 and 61.7 ± 25.1 μm, respectively, were significantly thinner than in HC eyes (P < 0.001 for both). However, in NMOSD–ON eyes, the GCIPL and pRNFL thicknesses were not significantly thinner than in HC eyes (P > 0.05 for both). In NMOSD+ON eyes, the RPCD and MSVD, 37.8 ± 7.1 and 36.7 ± 5.0%, respectively, were significantly less dense than HC eyes (P < 0.001 for both). Similarly, the RPCD and MSVD in NMOSD–ON eyes, 49.0 ± 2.8 and 43.9 ± 4.2%, respectively, were also less dense than in HC eyes (P < 0.029 for RPCD, P < 0.023 for MSVD). The highest AROC, 0.845 (sensitivity = 88.5%, specificity = 78.0%), was achieved by the logistic regression combination of all of the variables, i.e., pRNFL, GCIPL, RPCD, and MSVD.

Conclusions: Retinal microvascular changes were present in NMOSD–ON eyes. The combination of retinal structural and microvascular parameters might be helpful to discriminate NMOSD–ON eyes from HC eyes.

Universal digital filtering for denoising volumetric retinal OCT and OCT angiography in 3D shearlet domain

Retinal optical coherence tomography (OCT) and OCT angiography (OCTA) suffer from the degeneration of image quality due to speckle noise and bulk-motion noise, respectively. Because the cross-sectional retina has distinct features in OCT and OCTA B-scans, existing digital filters that can denoise OCT efficiently are unable to handle the bulk-motion noise in OCTA. In this Letter, we propose a universal digital filtering approach that is capable of minimizing both types of noise. Considering that the retinal capillaries in OCTA are hard to differentiate in B-scans while having distinct curvilinear structures in 3D volumes, we decompose the volumetric OCT and OCTA data with 3D shearlets, thus efficiently separating the retinal tissue and vessels from the noise in this transform domain. Compared with wavelets and curvelets, the shearlets provide better representation of the layer edges in OCT and the vasculature in OCTA. Qualitative and quantitative results show the proposed method outperforms the state-of-the-art OCT and OCTA denoising methods. Also, the superiority of 3D denoising is demonstrated by comparing the 3D shearlet filtering with its 2D counterpart.

Multi-resolution convolutional neural networks for fully automated segmentation of acutely injured lungs in multiple species

Segmentation of lungs with acute respiratory distress syndrome (ARDS) is a challenging task due to diffuse opacification in dependent regions which results in little to no contrast at the lung boundary. For segmentation of severely injured lungs, local intensity and texture information, as well as global contextual information, are important factors for consistent inclusion of intrapulmonary structures. In this study, we propose a deep learning framework which uses a novel multi-resolution convolutional neural network (ConvNet) for automated segmentation of lungs in multiple mammalian species with injury models similar to ARDS. The multi-resolution model eliminates the need to tradeoff between high-resolution and global context by using a cascade of low-resolution to high-resolution networks. Transfer learning is used to accommodate the limited number of training datasets. The model was initially pre-trained on human CT images, and subsequently fine-tuned on canine, porcine, and ovine CT images with lung injuries similar to ARDS. The multi-resolution model was compared to both high-resolution and low-resolution networks alone. The multi-resolution model outperformed both the low- and high-resolution models, achieving an overall mean Jacaard index of 0.963 ± 0.025 compared to 0.919 ± 0.027 and 0.950 ± 0.036, respectively, for the animal dataset (N=287). The multi-resolution model achieves an overall average symmetric surface distance of 0.438 ± 0.315 mm, compared to 0.971 ± 0.368 mm and 0.657 ± 0.519 mm for the low-resolution and high-resolution models, respectively. We conclude that the multi-resolution model produces accurate segmentations in severely injured lungs, which is attributed to the inclusion of both local and global features.

Depthwise Separable Convolutional Neural Network Model for Intra-Retinal Cyst Segmentation

Intra-retinal cysts (IRCs) are significant in detecting several ocular and retinal pathologies. Segmentation and quantification of IRCs from optical coherence tomography (OCT) scans is a challenging task due to present of speckle noise and scan intensity variations across the vendors. This work proposes a convolutional neural network (CNN) model with an encoder-decoder pair architecture for IRC segmentation across different cross-vendor OCT scans. Since deep CNN models have high computational complexity due to a large number of parameters, the proposed method of depthwise separable convolutional filters aids model generalizability and prevents model over-fitting. Also, the swish activation function is employed to prevent the vanishing gradient problem. The optima cyst segmentation challenge (OCSC) dataset with four different vendor OCT device scans is used to evaluate the proposed model. Our model achieves a mean Dice score of 0.74 and mean recall/precision rate of 0.72/0.82 across different imaging vendors and it outperforms existing algorithms on the OCSC dataset.

Is the Retina a Mirror of the Aging Brain? Aging of Neural Retina Layers and Primary Visual Cortex Across the Lifespan

How aging concomitantly modulates the structural integrity of the brain and retina in healthy individuals remains an outstanding question. Given the strong bottom-up retinocortical connectivity, it is important to study how these structures co-evolve during healthy aging in order to unravel mechanisms that may affect the physiological integrity of both structures. For the 56 participants in the study, primary visual cortex (BA17), as well as frontal, parietal and temporal regions thicknesses were measured in T1-weighted magnetic resonance imaging (MRI), and retinal macular thickness (10 neuroretinal layers) was measured by optical coherence tomography (OCT) imaging. We investigated the statistical association of these measures and their age dependence. We found an age-related decay of primary visual cortical thickness that was significantly correlated with a decrease in global and multiple layer retinal thicknesses. The atrophy of both structures might jointly account for the decline of various visual capacities that accompany the aging process. Furthermore, associations with other cortical regions suggest that retinal status may index cortical integrity in general.

Techniques and Applications in Skin OCT Analysis

The skin is the largest organ of our body. Skin disease abnormalities which occur within the skin layers are difficult to examine visually and often require biopsies to make a confirmation on a suspected condition. Such invasive methods are not well-accepted by children and women due to the possibility of scarring. Optical coherence tomography (OCT) is a non-invasive technique enabling in vivo examination of sub-surface skin tissue without the need for excision of tissue. However, one of the challenges in OCT imaging is the interpretation and analysis of OCT images. In this review, we discuss the various methodologies in skin layer segmentation and how it could potentially improve the management of skin diseases. We also present a review of works which use advanced machine learning techniques to achieve layers segmentation and detection of skin diseases. Lastly, current challenges in analysis and applications are also discussed.

Age-related changes of individual macular retinal layers among Asians

We characterized the age-related changes of the intra-retinal layers measured with spectral-domain optical coherence tomography (SD-OCT; Cirrus high-definition OCT [Carl Zeiss Meditec]. The Singapore Epidemiology of Eye Diseases is a population-based, cross-sectional study of Chinese, Malays and Indians living in Singapore. Iowa Reference Algorithms (Iowa Institute for Biomedical Imaging) were used for intra-retinal layer segmentation and mean thickness of 10 intra-retinal layers rescaled with magnification correction using axial length value. Linear regression models were performed to investigate the association of retinal layers with risk factors. After excluding participants with history of diabetes or ocular diseases, high-quality macular SD-OCT images were available for 2,047 participants (44–89 years old). Most of the retinal layers decreased with age except for foveal retinal nerve fiber layer (RNFL) and the inner/outer segments of photoreceptors where they increased with age. Men generally had thicker retinal layers than women. Chinese have the thickest RNFL and retinal pigment epithelium amongst the ethnic groups. Axial length and refractive error remained correlated with retinal layers in spite of magnification correction. Our data show pronounced age-related changes in retinal morphology. Age, gender, ethnicity and axial length need be considered when establishing OCT imaging biomarkers for ocular or systemic disease.

Accurate tissue interface segmentation via adversarial pre-segmentation of anterior segment OCT images

Optical Coherence Tomography (OCT) is an imaging modality that has been widely adopted for visualizing corneal, retinal and limbal tissue structure with micron resolution. It can be used to diagnose pathological conditions of the eye, and for developing pre-operative surgical plans. In contrast to the posterior retina, imaging the anterior tissue structures, such as the limbus and cornea, results in B-scans that exhibit increased speckle noise patterns and imaging artifacts. These artifacts, such as shadowing and specularity, pose a challenge during the analysis of the acquired volumes as they substantially obfuscate the location of tissue interfaces. To deal with the artifacts and speckle noise patterns and accurately segment the shallowest tissue interface, we propose a cascaded neural network framework, which comprises of a conditional Generative Adversarial Network (cGAN) and a Tissue Interface Segmentation Network (TISN). The cGAN pre-segments OCT B-scans by removing undesired specular artifacts and speckle noise patterns just above the shallowest tissue interface, and the TISN combines the original OCT image with the pre-segmentation to segment the shallowest interface. We show the applicability of the cascaded framework to corneal datasets, demonstrate that it precisely segments the shallowest corneal interface, and also show its generalization capacity to limbal datasets. We also propose a hybrid framework, wherein the cGAN pre-segmentation is passed to a traditional image analysis-based segmentation algorithm, and describe the improved segmentation performance. To the best of our knowledge, this is the first approach to remove severe specular artifacts and speckle noise patterns (prior to the shallowest interface) that affects the interpretation of anterior segment OCT datasets, thereby resulting in the accurate segmentation of the shallowest tissue interface. To the best of our knowledge, this is the first work to show the potential of incorporating a cGAN into larger deep learning frameworks for improved corneal and limbal OCT image segmentation. Our cGAN design directly improves the visualization of corneal and limbal OCT images from OCT scanners, and improves the performance of current OCT segmentation algorithms.

Increasing Accuracy of Optimal Surfaces Using Min-Marginal Energies

Optimal surface methods are a class of graph cut methods posing surface estimation as an n-ary ordered labeling problem. They are used in medical imaging to find interacting and layered surfaces optimally and in low order polynomial time. Representing continuous surfaces with discrete sets of labels, however, leads to discretization errors and, if graph representations are made dense, excessive memory usage. Limiting memory usage and computation time of graph cut methods are important and graphs that locally adapt to the problem has been proposed as a solution. Min-marginal energies computed using dynamic graph cuts offer a way to estimate solution uncertainty and these uncertainties have been used to decide where graphs should be adapted. Adaptive graphs, however, introduce extra parameters, complexity, and heuristics. We propose a way to use min-marginal energies to estimate continuous solution labels that does not introduce extra parameters and show empirically on synthetic and medical imaging datasets that it leads to improved accuracy. The increase in accuracy was consistent and in many cases comparable with accuracy otherwise obtained with graphs up to eight times denser, but with proportionally less memory usage and improvements in computation time.

Deep Learning Based Sub-Retinal Fluid Segmentation in Central Serous Chorioretinopathy Optical Coherence Tomography Scans

Development of an automated sub-retinal fluid segmentation technique from optical coherence tomography (OCT) scans is faced with challenges such as noise and motion artifacts present in OCT images, variation in size, shape and location of fluid pockets within the retina. The ability of a fully convolutional neural network to automatically learn significant low level features to differentiate subtle spatial variations makes it suitable for retinal fluid segmentation task. Hence, a fully convolutional neural network has been proposed in this work for the automatic segmentation of sub-retinal fluid in OCT scans of central serous chorioretinopathy (CSC) pathology. The proposed method has been evaluated on a dataset of 15 OCT volumes and an average Dice rate, Precision and Recall of 0.91, 0.93 and 0.89 respectively has been achieved over the test set.

Coupled s-excess HMM for vessel border tracking and segmentation

In this paper, we present a novel image segmentation technique, based on hidden Markov model (HMM), which we then apply to simultaneously segment interior and exterior walls of fluorescent confocal images of lymphatic vessels. Our proposed method achieves this by tracking hidden states, which are used to indicate the locations of both the inner and outer wall borders throughout the sequence of images. We parameterize these vessel borders using radial basis functions (RBFs), thus enabling us to minimize the number of points we need to track as we progress through multiple layers and therefore reduce computational complexity. Information about each border is detected using patch-wise convolutional neural networks (CNN). We use the softmax function to infer the emission probability and use a proposed new training algorithm based on s-excess optimization to learn the transition probability. We also introduce a new optimization method to determine the optimum sequence of the hidden states. Thus, we transform the segmentation problem into one that minimizes an s-excess graph cut, where each hidden state is represented as a graph node and the weight of these nodes are defined by their emission probabilities. The transition probabilities are used to define relationships between neighboring nodes in the constructed graph. We compare our proposed method to the Viterbi and Baum-Welch algorithms. Both qualitative and quantitative analysis show superior performance of the proposed methods.

Segmentation of mouse skin layers in optical coherence tomography image data using deep convolutional neural networks

Optical coherence tomography (OCT) enables the non-invasive acquisition of high-resolution three-dimensional cross-sectional images at micrometer scale and is mainly used in the field of ophthalmology for diagnosis as well as monitoring of eye diseases. Also in other areas, such as dermatology, OCT is already well established. Due to its non-invasive nature, OCT is also employed for research studies involving animal models. Manual evaluation of OCT images of animal models is a challenging task due to the lack of imaging standards and the varying anatomy among models. In this paper, we present a deep learning algorithm for the automatic segmentation of several layers of mouse skin in OCT image data using a deep convolutional neural network (CNN). The architecture of our CNN is based on the U-net and is modified by densely connected convolutions. We compared our adapted CNN with our previous algorithm, a combination of a random forest classification and a graph-based refinement, and a baseline U-net. The results showed that, on average, our proposed CNN outperformed our previous algorithm and the baseline U-net. In addition, a reduction of outliers could be observed through the use of densely connected convolutions.

Compensation of retinal nerve fibre layer thickness as assessed using optical coherence tomography based on anatomical confounders

Background/Aims

To compensate the retinal nerve fibre layer (RNFL) thickness assessed by spectral-domain optical coherence tomography (SD-OCT) for anatomical confounders.

Methods

The Singapore Epidemiology of Eye Diseases is a population-based study, where 2698 eyes (1076 Chinese, 704 Malays and 918 Indians) with high-quality SD-OCT images from individuals without eye diseases were identified. Optic disc and macular cube scans were registered to determine the distance between fovea and optic disc centres (fovea distance) and their respective angle (fovea angle). Retinal vessels were segmented in the projection images and used to calculate the circumpapillary retinal vessel density profile. Compensated RNFL thickness was generated based on optic disc (ratio, orientation and area), fovea (distance and angle), retinal vessel density, refractive error and age. Linear regression models were used to investigate the effects of clinical factors on RNFL thickness.

Results

Retinal vessel density reduced significantly with increasing age (1487±214 µm in 40–49, 1458±208 µm in 50–59, 1429±223 µm in 60–69 and 1415±233 µm in ≥70). Compensation reduced the variability of RNFL thickness, where the effect was greatest for Chinese (10.9%; p<0.001), followed by Malays (6.6%; p=0.075) and then Indians (4.3%; p=0.192). Compensation reduced the age-related RNFL decline by 55% in all participants (β=−3.32 µm vs β=−1.50 µm/10 years; p<0.001). Nearly 62% of the individuals who were initially classified as having abnormally thin RNFL (outside the 99% normal limits) were later reclassified as having normal RNFL.

Conclusions

RNFL thickness compensated for anatomical parameters reduced the variability of measurements and may improve glaucoma detection, which needs to be confirmed in future studies.

Persistent and Progressive Outer Retina Thinning in Frontotemporal Degeneration

OBJECTIVE

While Alzheimer's disease is associated with inner retina thinning measured by spectral-domain optical coherence tomography (SD-OCT), our previous cross-sectional study suggested outer retina thinning in frontotemporal degeneration (FTD) patients compared to controls without neurodegenerative disease; we sought to evaluate longitudinal changes of this potential biomarker.

METHODS

SD-OCT retinal layer thicknesses were measured at baseline and after 1-2 years. Clinical criteria, genetic analysis, and a cerebrospinal fluid biomarker (total tau: β-amyloid) to exclude likely underlying Alzheimer's disease pathology were used to define a subgroup of predicted molecular pathology (i.e., tauopathy). Retinal layer thicknesses and rates of change in all FTD patients (n = 16 patients, 30 eyes) and the tauopathy subgroup (n = 9 patients,16 eyes) were compared to controls (n = 30 controls, 47 eyes) using a generalized linear model accounting for inter-eye correlation and adjusting for age, sex, and race. Correlations between retinal layer thicknesses and Mini-Mental State Examinations (MMSE) were assessed.

RESULTS

Compared to controls, returning FTD patients (143 vs. 130 μm, p = 0.005) and the tauopathy subgroup (143 vs. 128 μm, p = 0.03) had thinner outer retinas but similar inner layer thicknesses. Compared to controls, the outer retina thinning rate was not significant for all FTD patients (p = 0.34), but was significant for the tauopathy subgroup (-3.9 vs. 0.4 μm/year, p = 0.03). Outer retina thickness change correlated with MMSE change in FTD patients (Spearman rho = 0.60, p = 0.02) and the tauopathy subgroup (rho = 0.73, p = 0.04).

CONCLUSION

Our finding of FTD outer retina thinning persists and longitudinally correlates with disease progression. These findings were especially seen in probable tauopathy patients, which showed progressive outer retina thinning.

Prediction of Individual Disease Conversion in Early AMD Using Artificial Intelligence

Purpose

While millions of individuals show early age-related macular degeneration (AMD) signs, yet have excellent vision, the risk of progression to advanced AMD with legal blindness is highly variable. We suggest means of artificial intelligence to individually predict AMD progression.

Methods

In eyes with intermediate AMD, progression to the neovascular type with choroidal neovascularization (CNV) or the dry type with geographic atrophy (GA) was diagnosed based on standardized monthly optical coherence tomography (OCT) images by independent graders. We obtained automated volumetric segmentation of outer neurosensory layers and retinal pigment epithelium, drusen, and hyperreflective foci by spectral domain-OCT image analysis. Using imaging, demographic, and genetic input features, we developed and validated a machine learning-based predictive model assessing the risk of conversion to advanced AMD.

Results

Of a total of 495 eyes, 159 eyes (32%) had converted to advanced AMD within 2 years, 114 eyes progressed to CNV, and 45 to GA. Our predictive model differentiated converting versus nonconverting eyes with a performance of 0.68 and 0.80 for CNV and GA, respectively. The most critical quantitative features for progression were outer retinal thickness, hyperreflective foci, and drusen area. The features for conversion showed pathognomonic patterns that were distinctly different for the neovascular and the atrophic pathways. Predictive hallmarks for CNV were mostly drusen-centric, while GA markers were associated with neurosensory retina and age.

Conclusions

Artificial intelligence with automated analysis of imaging biomarkers allows personalized prediction of AMD progression. Moreover, pathways of progression may be specific in respect to the neovascular/atrophic type.

Layer boundary evolution method for macular OCT layer segmentation

Optical coherence tomography (OCT) is used to produce high resolution depth images of the retina and is now the standard of care for in-vivo ophthalmological assessment. It is also increasingly being used for evaluation of neurological disorders such as multiple sclerosis (MS). Automatic segmentation methods identify the retinal layers of the macular cube providing consistent results without intra- and inter-rater variation and is faster than manual segmentation. In this paper, we propose a fast multi-layer macular OCT segmentation method based on a fast level set method. Our framework uses contours in an optimized approach specifically for OCT layer segmentation over the whole macular cube. Our algorithm takes boundary probability maps from a trained random forest and iteratively refines the prediction to subvoxel precision. Evaluation on both healthy and multiple sclerosis subjects shows that our method is statistically better than a state-of-the-art graph-based method.

Muscle fibre morphology and microarchitecture in cerebral palsy patients obtained by 3D synchrotron X-ray computed tomography

BACKGROUND

Synchrotron X-ray computed tomography (SXCT) allows for three-dimensional imaging of objects at a very high resolution and in large field-of-view.

PURPOSE

The aim of this study was to use SXCT imaging for morphological analysis of muscle tissue, in order to investigate whether the analysis reveals complementary information to two-dimensional microscopy.

METHODS

Three-dimensional SXCT images of muscle biopsies were taken from participants with cerebral palsy and from healthy controls. We designed morphological measures from the two-dimensional slices and three-dimensional volumes of the images and measured the muscle fibre organization, which we term orientation consistency.

RESULTS

The muscle fibre cross-sectional areas were significantly larger in healthy participants than in participants with cerebral palsy when carrying out the analysis in three dimensions. However, a similar analysis carried out in two dimensions revealed no patient group difference. The present study also showed that three-dimensional orientation consistency was significantly larger for healthy participants than for participants with cerebral palsy.

CONCLUSION

Individuals with CP have smaller muscle fibres than healthy control individuals. We argue that morphometric measures of muscle fibres in two dimensions are generally trustworthy only if the fibres extend perpendicularly to the slice plane, and otherwise three-dimensional aspects should be considered. In addition, the muscle tissue of individuals with CP showed a decreased level of orientation consistency when compared to healthy control tissue. We suggest that the observed disorganization of the tissue may be induced by atrophy caused by physical inactivity and insufficient neural activation.

Optimal surface segmentation with convex priors in irregularly sampled space

Optimal surface segmentation is a state-of-the-art method used for segmentation of multiple globally optimal surfaces in volumetric datasets. The method is widely used in numerous medical image segmentation applications. However, nodes in the graph based optimal surface segmentation method typically encode uniformly distributed orthogonal voxels of the volume. Thus the segmentation cannot attain an accuracy greater than a single unit voxel, i.e. the distance between two adjoining nodes in graph space. Segmentation accuracy higher than a unit voxel is achievable by exploiting partial volume information in the voxels which shall result in non-equidistant spacing between adjoining graph nodes. This paper reports a generalized graph based multiple surface segmentation method with convex priors which can optimally segment the target surfaces in an irregularly sampled space. The proposed method allows non-equidistant spacing between the adjoining graph nodes to achieve subvoxel segmentation accuracy by utilizing the partial volume information in the voxels. The partial volume information in the voxels is exploited by computing a displacement field from the original volume data to identify the subvoxel-accurate centers within each voxel resulting in non-equidistant spacing between the adjoining graph nodes. The smoothness of each surface modeled as a convex constraint governs the connectivity and regularity of the surface. We employ an edge-based graph representation to incorporate the necessary constraints and the globally optimal solution is obtained by computing a minimum s-t cut. The proposed method was validated on 10 intravascular multi-frame ultrasound image datasets for subvoxel segmentation accuracy. In all cases, the approach yielded highly accurate results. Our approach can be readily extended to higher-dimensional segmentations.

3D segmentation of exterior wall surface of abdominal aortic aneurysm from CT images using variable neighborhood search

A 3D model of abdominal aortic aneurysm (AAA) can provide useful anatomical information for clinical management and simulation. Thin-slice contiguous computed tomographic (CT) angiography is the best source of medical images for construction of 3D models, which requires segmentation of AAA in the images. Existing methods for segmentation of AAA rely on either manual process or 2D segmentation in each 2D CT slide. However, a traditional manual segmentation is a time consuming process which is not practical for routine use. The construction of a 3D model from 2D segmentation of each CT slice is not a fully satisfactory solution due to rough contours that can occur because of lack of constraints among segmented slices, as well as missed segmentation slices. To overcome such challenges, this paper proposes the 3D segmentation of AAA using the concept of variable neighborhood search by iteratively alternating between two different segmentation techniques in the two different 3D search spaces of voxel intensity and voxel gradient. The segmentation output of each method is used as the initial contour to the other method in each iteration. By alternating between search spaces, the technique can escape local minima that naturally occur in each search space. Also, the 3D search spaces provide more constraints across CT slices, when compared with the 2D search spaces in individual CT slices. The proposed method is evaluated with 10 easy and 10 difficult cases of AAA. The results show that the proposed 3D segmentation technique achieves the outstanding segmentation accuracy with an average dice similarity value (DSC) of 91.88%, when compared to the other methods using the same dataset, which are the 2D proposed method, classical graph cut, distance regularized level set evolution, and registration based geometric active contour with the DSCs of 87.57 ± 4.52%, 72.47 ± 8.11%, 58.50 ± 8.86% and 76.21 ± 10.49%, respectively.

Automatic three-dimensional segmentation of endoscopic airway OCT images

Automatic delineation and segmentation of airway structures from endoscopic optical coherence tomography (OCT) images improve image analysis efficiency and thus has been of particular interest. Conventional two-dimensional automatic segmentation methods, such as the dynamic programming approach, ensures the edge-continuity in the xz-direction (intra-B-scan), but fails to preserve the surface-continuity when concerning the y-direction (inter-B-scan). To solve this, we present a novel automatic three-dimensional (3D) airway segmentation strategy. Our segmentation scheme includes an artifact-oriented pre-processing pipeline and a modified 3D optimal graph search algorithm incorporating adaptive tissue-curvature adjustment. The proposed algorithm is tested on endoscopic airway OCT image data sets acquired by different swept-source OCT platforms, and on different animal and human models. With our method, the results show continuous surface segmentation performance, which is both robust and accurate.

Interplay Between Macular Retinal Changes and White Matter Integrity in Early Alzheimer's Disease

This study aims to investigate the relationship between structural changes in the retina and white matter in the brain, in early Alzheimer's disease (AD). Twenty-three healthy controls (mean age = 63.4±7.5 years) and seventeen AD patients (mean age = 66.5±6.6 years) were recruited for this study. By combining two imaging techniques-optical coherence tomography and diffusion tensor imaging (DTI)-the association between changes in the thickness of individual retinal layers and white matter dysfunction in early AD was assessed. Retinal layers were segmented, and thickness measurements were obtained for each layer. DTI images were analyzed with a quantitative data-driven approach to evaluating whole-brain diffusion metrics, using tract-based spatial statistics. Diffusion metrics, such as fractional anisotropy, are markers for white matter integrity. Multivariate and partial correlation analyses evaluating the association between individual retinal layers thickness and diffusion metrics were performed. We found that axial diffusivity, indexing axonal integrity, was significantly reduced in AD (p = 0.016, Cohen's d = 1.004) while in the retina, only a marginal trend for significance was found for the outer plexiform layer (p = 0.084, Cohen's d = 0.688). Furthermore, a positive association was found in the AD group between fractional anisotropy and the inner nuclear layer thickness (p < 0.05, r = 0.419, corrected for multiple comparisons by controlling family-wise error rate). Our findings suggest that axonal damage in the brain dominates early on in this condition and shows an association with retinal structural integrity already at initial stages of AD. These findings are consistent with an early axonal degeneration mechanism in AD.

Quantitative 3D Analysis of Coronary Wall Morphology in Heart Transplant Patients: OCT-Assessed Cardiac Allograft Vasculopathy Progression

Cardiac allograft vasculopathy (CAV) accounts for about 30% of all heart-transplant (HTx) patient deaths. For patients at high risk for CAV complications after HTx, therapy must be initiated early to be effective. Therefore, new phenotyping approaches are needed to identify such HTx patients at the earliest possible time. Coronary optical coherence tomography (OCT) images were acquired from 50 HTx patients 1 and 12 months after HTx. Quantitative analysis of coronary wall morphology used LOGISMOS segmentation strategy to simultaneously identify three wall-layer surfaces for the entire pullback length in 3D: luminal, outer intimal, and outer medial surfaces. To quantify changes of coronary wall morphology between 1 and 12 months after HTx, the two pullbacks were mutually co-registered. Validation of layer thickness measurements showed high accuracy of performed layer analyses with layer thickness measures correlating well with manually-defined independent standard (Rautomated2 = 0.93, y=1.0x-6.2μm), average intimal+medial thickness errors were 4.98 ± 31.24 µm, comparable with inter-observer variability. Quantitative indices of coronary wall morphology 1 month and 12 months after HTx showed significant local as well as regional changes associated with CAV progression. Some of the newly available fully-3D baseline indices (intimal layer brightness, medial layer brightness, medial thickness, and intimal+medial thickness) were associated with CAV-related progression of intimal thickness showing promise of identifying patients subjected to rapid intimal thickening at 12 months after HTx from OCT-image data obtained just 1 month after HTx. Our approach allows quantification of location-specific alterations of coronary wall morphology over time and is sensitive even to very small changes of wall layer thicknesses that occur in patients following heart transplant.