Faculty Bibliography

PURPOSE: To develop an automated method to identify the normal macula and three macular pathologies (macular hole [MH], macular edema [ME], and age-related macular degeneration [AMD]) from the fovea-centered cross sections in three-dimensional (3D) spectral-domain optical coherence tomography (SD-OCT) images. METHODS: A sample of SD-OCT macular scans (macular cube 200 x 200 or 512 x 128 scan protocol; Cirrus HD-OCT; Carl Zeiss Meditec, Inc., Dublin, CA) was obtained from healthy subjects and subjects with MH, ME, and/or AMD (dataset for development: 326 scans from 136 subjects [193 eyes], and dataset for testing: 131 scans from 37 subjects [58 eyes]). A fovea-centered cross-sectional slice for each of the SD-OCT images was encoded using spatially distributed multiscale texture and shape features. Three ophthalmologists labeled each fovea-centered slice independently, and the majority opinion for each pathology was used as the ground truth. Machine learning algorithms were used to identify the discriminative features automatically. Two-class support vector machine classifiers were trained to identify the presence of normal macula and each of the three pathologies separately. The area under the receiver operating characteristic curve (AUC) was calculated to assess the performance. RESULTS: The cross-validation AUC result on the development dataset was 0.976, 0.931, 0939, and 0.938, and the AUC result on the holdout testing set was 0.978, 0.969, 0.941, and 0.975, for identifying normal macula, MH, ME, and AMD, respectively. CONCLUSIONS: The proposed automated data-driven method successfully identified various macular pathologies (all AUC > 0.94). This method may effectively identify the discriminative features without relying on a potentially error-prone segmentation module.

We address a novel problem domain in the analysis of optical coherence tomography (OCT) images: the diagnosis of multiple macular pathologies in retinal OCT images. The goal is to identify the presence of normal macula and each of three types of macular pathologies, namely, macular edema, macular hole, and age-related macular degeneration, in the OCT slice centered at the fovea. We use a machine learning approach based on global image descriptors formed from a multi-scale spatial pyramid. Our local features are dimension-reduced local binary pattern histograms, which are capable of encoding texture and shape information in retinal OCT images and their edge maps, respectively. Our representation operates at multiple spatial scales and granularities, leading to robust performance. We use 2-class support vector machine classifiers to identify the presence of normal macula and each of the three pathologies. To further discriminate sub-types within a pathology, we also build a classifier to differentiate full-thickness holes from pseudo-holes within the macular hole category. We conduct extensive experiments on a large dataset of 326 OCT scans from 136 subjects. The results show that the proposed method is very effective (all AUC>0.93).

Aqueous humor (AH) exiting the eye via the trabecular meshwork and Schlemm's canal (SC) passes through the deep and intrascleral venous plexus (ISVP) or directly through aqueous veins. The purpose of this study was to visualize the human AH outflow system 360 degrees in three dimensions (3D) during active AH outflow in a virtual casting. The conventional AH outflow pathways of 7 donor eyes were imaged with a modified Bioptigen spectral-domain optical coherence tomography system (Bioptigen Inc, USA; SuperLum LTD, Ireland) at a perfusion pressure of 20 mmHg (N = 3), and 10 mmHg (N = 4). In all eyes, 36 scans (3 equally distributed in each clock hour), each covering a 2 x 3 x 2 mm volume (512 frames, each 512 x 1024 pixels), were obtained. All image data were black/white inverted, and the background subtracted (ImageJ 1.40 g, http://rsb.info.nih.gov/ij/). Contrast was adjusted to isolate the ISVP. SC, collector channels, the deep and ISVP, and episcleral veins were observed throughout the limbus. Aqueous veins could be observed extending into the episcleral veins. Individual scan ISVP castings were rendered and assembled in 3D space in Amira 4.1 (Visage Imaging Inc. USA). A 360-degree casting of the ISVP was obtained in all perfused eyes. The ISVP tended to be dense and overlapping in the superior and inferior quadrants, and thinner in the lateral quadrants. The human AH outflow pathway can be imaged using SD-OCT. The more superficial structures of the AH outflow pathway present with sufficient contrast as to be optically isolated and cast in-situ 360 degrees in cadaver eye perfusion models. This approach may be useful as a model in future studies of human AH outflow.

Optical coherence tomography (OCT) techniques have been applied to develop a new generation of the technology, called spectral domain (SD) or Fourier domain (FD) OCT. The commercially available SD-OCT technology offers benefits over the conventional time domain (TD) OCT such as a scanning speed up to 200 times faster and higher axial resolution (3 to 6 mum). Overall, SD-OCT offers improved performance in terms of reproducibility. SD-OCT has a level of discriminating capability, between healthy and perimetric glaucoma eyes similar to that obtained with TD-OCT. Furthermore, the capabilities and features of SD-OCT are rapidly evolving, mainly due to three-dimensional imaging and image rendering. More sophisticated approaches for macular and optic disc assessment are expected to be employed in clinical practice. Analysis software should be further refined for interpretation of SD-OCT images in order to enhance the sensitivity and specificity of glaucoma diagnostics. Most importantly for SD-OCT is determination of its ability to diagnostic structural glaucomatous progression. Considering the recent launch time of the commercially available SD-OCT and slow progressing characteristic of glaucoma, we must wait for longitudinal SD-OCT data, with a long enough follow-up, to become available.

Detection of disease progression is an important and challenging component of glaucoma management. Optical coherence tomography (OCT) has proved to be valuable in the detection of glaucomatous damage. With its high resolution and proven measurement reproducibility, OCT has the potential to become an important tool for glaucoma progression detection. This manuscript presents the capabilities of the OCT technology pertinent for detection of progressive glaucomatous damage and provides a review of the current knowledge on the device's clinical performance.

Optical coherence tomography (OCT) imaging has become widespread in ophthalmology over the past 15 years, because of its ability to visualize ocular structures at high resolution. This article reviews the history of OCT imaging of the eye, its current status, and the laboratory work that is driving the future of the technology.

PURPOSE: To investigate the longitudinal effect of optic nerve crush injury in mice by measuring retinal thickness with spectral-domain optical coherence tomography (SD-OCT). METHODS: Optic nerves of one eye from each C57Bl/6 mouse were crushed under direct visualization for 3 seconds, 1 mm posterior to the globe. The optic nerve head (ONH) was imaged with SD-OCT (1.5 x 1.5 x 2.0 mm scan) before the surgical intervention and repeated subsequently for up to 32 days postinjury. A cohort of mice not exposed to the nerve crush procedure served as control. En face SD-OCT images were used to manually align subsequent scans to the baseline en face image. Total retinal thickness (TRT) (along a sampling band with radii 0.33-0.42 mm centered on the ONH) from each follow-up day was automatically quantified for global and sectoral measurements using custom software. Linear mixed-effects models with quadratic terms were fitted to compare TRT of nerve-crushed and control eyes over time. RESULTS: Eleven eyes from 11 nerve crush mice (baseline age 76 +/- 11.8 days) and eight eyes from four healthy mice (baseline age 64 +/- 0 days) were included. The control eyes showed a small, gradual, and consistent TRT increase throughout follow-up. Nerve-crushed eyes showed an initial period of thickening, followed by thinning and slight rebound after day 21. The decrease in thickness observed after the early thickening resolved was statistically significantly different from the control eyes (P < 0.05 for global and sectoral measurements). CONCLUSIONS: SD-OCT can be used to quantitatively monitor changes in retinal thickness in mice over time.

Current standard quantitative 3D spectral-domain optical coherence tomography (SD-OCT) analyses of various ocular diseases is limited in detecting structural damage at early pathologic stages. This is mostly because only a small fraction of the 3D data is used in the current method of quantifying the structure of interest. This paper presents a novel SD-OCT data analysis technique, taking full advantage of the 3D dataset. The proposed algorithm uses machine classifier to analyze SD-OCT images after grouping adjacent pixels into super pixel in order to detect glaucomatous damage. A 3D SD-OCT image is first converted into a 2D feature map and partitioned into over a hundred super pixels. Machine classifier analysis using boosting algorithm is performed on super pixel features. One hundred and ninety-two 3D OCT images of the optic nerve head region were tested. Area under the receiver operating characteristic (AUC) was computed to evaluate the glaucoma discrimination performance of the algorithm and compare it to the commercial software output. The AUC of normal vs glaucoma suspect eyes using the proposed method was statistically significantly higher than the current method (0.855 and 0.707, respectively, p=0.031). This new method has the potential to improve early detection of glaucomatous structural damages.

Purpose. To describe the profile and identify the predictors of the ganglion cell-inner plexiform layer (GCIPL) thickness measured with frequency-domain optical coherence tomography (FD-OCT) in normal eyes. Methods. Two hundred eighty-two normal subjects underwent macular and optic disc scanning in both eyes with Cirrus high-definition (HD)-OCT (Carl Zeiss Meditec, Dublin, CA). Linear regression analyses were performed to determine the association between GCIPL thickness and age, sex, ethnicity (Europeans, Africans, Hispanics, Asians, and Indians), eye laterality, refraction, intraocular pressure, axial length, central corneal thickness, mean retinal nerve fiber layer (RNFL) thickness, disc and rim areas, cup-to-disc area, vertical and horizontal cup-to-disc diameter ratios, vertical rim thickness, and OCT signal strength. Results. The mean (+/-SD) age was 46.2 +/- 16.9 years (range, 18-84 years). The mean and minimum GCIPL thicknesses (+/-SD) were 82.1 +/- 6.2 and 80.4 +/- 6.4 mum, respectively. There were significant differences in GCIPL thickness between macular sectors (P < 0.05), except between the superotemporal and inferonasal sectors (P = 0.63). The superonasal sector had the thickest and the inferior had the thinnest GCIPL. The GCIPL of the superior hemisphere was thicker than that of the inferior, and the nasal sector GCIPL was significantly thicker than the temporal one (P < 0.001). The average GCIPL did not differ between male and female subjects (P = 0.16) after adjustment for axial length and between ethnic groups (P = 0.41) after adjustment for age, axial length, and RNFL thickness. Significant predictors of mean GCIPL thickness were average RNFL thickness (beta = 0.37, P < 0.001), age (beta = -0.083, P < 0.001), axial length (beta = -0.87, P = 0.001), and male sex (beta = -1.62, P = 0.005). Conclusions. The independent factors associated with thinner GCIPL include thinner RNFL, older age, longer ocular axial length, and being male. Although the magnitude of the effect of age, axial length, and sex are small, these factors should be taken into account when interpreting Cirrus HD-OCT-based GCIPL thickness measurements

PURPOSE: To test the reproducibility of spectral-domain optical coherence tomography (SD-OCT) total retinal thickness (TRT) measurements in mice. METHODS: C57Bl/6 mice were anesthetized, and three repeated volumetric images were acquired in both eyes with SD-OCT (250 A-scans x 250 frames x 1024 samplings), centered on the optic nerve head (ONH). The mice were repositioned between scans. TRT was automatically measured within a sampling band of retinal thickness with radii of 55 to 70 pixels, centered on the ONH by using custom segmentation software. The first volumetric image acquired in a given eye was used to register the remaining two SD-OCT images by manually aligning the en face images with respect to rotation and linear translation. Linear mixed-effects models were fitted to global and quadrant thicknesses, taking into account the clustering between eyes, to assess imprecision (measurement reproducibility). RESULTS: Twenty-six eyes of 13 adult mice (age 13 weeks) were imaged. The mean global TRT across all eyes was 298.21 mum, with a mouse heterogeneity standard deviation (SD) of 4.88 mum (coefficient of variation [CV] = 0.016), an eye SD of 3.32 mum (CV = 0.011), and a device-related imprecision SD of 2.33 mum (CV = 0.008). The superior quadrant had the thickest mean TRT measurement (310.38 mum) and the highest (worst) imprecision SD (3.13 mum; CV = 0.010), and the inferior quadrant had the thinnest mean TRT (291.55 mum). The quadrant with the lowest (best) imprecision SD was in the nasal one (2.06 mum; CV = 0.007). CONCLUSIONS: Good reproducibility was observed for SD-OCT retinal thickness measurements in mice. SD-OCT may be useful for in vivo longitudinal studies in mice.