Boglárka Varga

Real-Time Automatic Segmentation of Optical Coherence Tomography Volume Data of the Macular Region.

Optical coherence tomography (OCT) is a high speed, high resolution and non-invasive imaging modality that enables the capturing of the 3D structure of the retina. The fast and automatic analysis of 3D volume OCT data is crucial taking into account the increased amount of patient-specific 3D imaging data. In this work, we have developed an automatic algorithm, OCTRIMA 3D (OCT Retinal IMage Analysis 3D), that could segment OCT volume data in the macular region fast and accurately. The proposed method is implemented using the shortest-path based graph search, which detects the retinal boundaries by searching the shortest-path between two end nodes using Dijkstra’s algorithm. Additional techniques, such as inter-frame flattening, inter-frame search region refinement, masking and biasing were introduced to exploit the spatial dependency between adjacent frames for the reduction of the processing time. Our segmentation algorithm was evaluated by comparing with the manual labelings and three state of the art graph-based segmentation methods. The processing time for the whole OCT volume of 496×644×51 voxels (captured by Spectralis SD-OCT) was 26.15 seconds which is at least a 2-8-fold increase in speed compared to other, similar reference algorithms used in the comparisons. The average unsigned error was about 1 pixel (∼ 4 microns), which was also lower compared to the reference algorithms. We believe that OCTRIMA 3D is a leap forward towards achieving reliable, real-time analysis of 3D OCT retinal data.

Fractal-based analysis of optical coherence tomography data to quantify retinal tissue damage

BackgroundThe sensitivity of Optical Coherence Tomography (OCT) images to identify retinal tissue morphology characterized by early neural loss from normal healthy eyes is tested by calculating structural information and fractal dimension. OCT data from 74 healthy eyes and 43 eyes with type 1 diabetes mellitus with mild diabetic retinopathy (MDR) on biomicroscopy was analyzed using a custom-built algorithm (OCTRIMA) to measure locally the intraretinal layer thickness. A power spectrum method was used to calculate the fractal dimension in intraretinal regions of interest identified in the images. ANOVA followed by Newman-Keuls post-hoc analyses were used to test for differences between pathological and normal groups. A modified p value of <0.001 was considered statistically significant. Receiver operating characteristic (ROC) curves were constructed to describe the ability of each parameter to discriminate between eyes of pathological patients and normal healthy eyes.ResultsFractal dimension was higher for all the layers (except the GCL + IPL and INL) in MDR eyes compared to normal healthy eyes. When comparing MDR with normal healthy eyes, the highest AUROC values estimated for the fractal dimension were observed for GCL + IPL and INL. The maximum discrimination value for fractal dimension of 0.96 (standard error =0.025) for the GCL + IPL complex was obtained at a FD ≤ 1.66 (cut off point, asymptotic 95% Confidence Interval: lower-upper bound = 0.905-1.002). Moreover, the highest AUROC values estimated for the thickness measurements were observed for the OPL, GCL + IPL and OS. Particularly, when comparing MDR eyes with control healthy eyes, we found that the fractal dimension of the GCL + IPL complex was significantly better at diagnosing early DR, compared to the standard thickness measurement.ConclusionsOur results suggest that the GCL + IPL complex, OPL and OS are more susceptible to initial damage when comparing MDR with control healthy eyes. Fractal analysis provided a better sensitivity, offering a potential diagnostic predictor for detecting early neurodegeneration in the retina.

Automated classifiers for early detection and diagnosis of retinopathy in diabetic eyes

BackgroundArtificial neural networks (ANNs) have been used to classify eye diseases, such as diabetic retinopathy (DR) and glaucoma. DR is the leading cause of blindness in working-age adults in the developed world. The implementation of DR diagnostic routines could be feasibly improved by the integration of structural and optical property test measurements of the retinal structure that provide important and complementary information for reaching a diagnosis. In this study, we evaluate the capability of several structural and optical features (thickness, total reflectance and fractal dimension) of various intraretinal layers extracted from optical coherence tomography images to train a Bayesian ANN to discriminate between healthy and diabetic eyes with and with no mild retinopathy.ResultsWhen exploring the probability as to whether the subject’s eye was healthy (diagnostic condition, Test 1), we found that the structural and optical property features of the outer plexiform layer (OPL) and the complex formed by the ganglion cell and inner plexiform layers (GCL + IPL) provided the highest probability (positive predictive value (PPV) of 91% and 89%, respectively) for the proportion of patients with positive test results (healthy condition) who were correctly diagnosed (Test 1). The true negative, TP and PPV values remained stable despite the different sizes of training data sets (Test 2). The sensitivity, specificity and PPV were greater or close to 0.70 for the retinal nerve fiber layer’s features, photoreceptor outer segments and retinal pigment epithelium when 23 diabetic eyes with mild retinopathy were mixed with 38 diabetic eyes with no retinopathy (Test 3).ConclusionsA Bayesian ANN trained on structural and optical features from optical coherence tomography data can successfully discriminate between healthy and diabetic eyes with and with no retinopathy. The fractal dimension of the OPL and the GCL + IPL complex predicted by the Bayesian radial basis function network provides better diagnostic utility to classify diabetic eyes with mild retinopathy. Moreover, the thickness and fractal dimension parameters of the retinal nerve fiber layer, photoreceptor outer segments and retinal pigment epithelium show promise for the diagnostic classification between diabetic eyes with and with no mild retinopathy.

Performance evaluation of automated segmentation software on optical coherence tomography volume data

Over the past two decades a significant number of OCT segmentation approaches have been proposed in the literature. Each methodology has been conceived for and/or evaluated using specific datasets that do not reflect the complexities of the majority of widely available retinal features observed in clinical settings. In addition, there does not exist an appropriate OCT dataset with ground truth that reflects the realities of everyday retinal features observed in clinical settings. While the need for unbiased performance evaluation of automated segmentation algorithms is obvious, the validation process of segmentation algorithms have been usually performed by comparing with manual labelings from each study and there has been a lack of common ground truth. Therefore, a performance comparison of different algorithms using the same ground truth has never been performed. This paper reviews research-oriented tools for automated segmentation of the retinal tissue on OCT images. It also evaluates and compares the performance of these software tools with a common ground truth.

Identifying Local Structural and Optical Derangement in the Neural Retina of Individuals with Type 1 Diabetes

Background: To identify local structural and optical derangement in the neural retina of individuals with type 1 diabetes having early diabetic retinopathy (DR) and compare with healthy non-diabetic controls and type 1 diabetic individuals having no DR. Methods: Optical coherence tomography (TDOCT) examination was performed on a total of 74 healthy eyes, 38 eyes with type 1 diabetes mellitus (DM) with no retinopathy and 43 eyes with mild DR (MDR). A total of 6 intraretinal layers were segmented on OCT images. Thickness and reflectance-based measurements were extracted for each OCT scan using features measured locally for each intraretinal layer. Results: In the analysis where local measures were averaged in the separated macular regions outside the foveola, the mean thickness values of the outer segment of photoreceptors (OS) in the perifoveal region, ganglion cell and inner plexiform layer (GCL+IPL) complex in the parafoveal and outer plexiform layer (OPL) in the foveal region were significantly smaller (13%, 8% and 36%; respectively, p<0.001) when comparing MDR eyes with controls. The mean thickness values of the OPL (foveal region, 27%, p<0.001) and the OS (parafoveal (24%) and perifoveal (23%), p<0.001) were significantly smaller when comparing MDR with DM eyes. The reflectance-based measures were significantly smaller for all layers in MDR eyes compared with healthy and DM eyes (7-36%, p<0.001). Conclusions: Our results show OCT is capable of detecting selective layer thinning and that the optical properties extracted from OCT images add significant evidence to the morphological information directly provided by OCT. It also suggests that the outer segment of the photoreceptor layer may be vulnerable in both type 1 diabetic individuals with and without early DR. Our results might also indicate that an early sign of vascular alteration development could be detected by investigating the changes in optical properties and thickness of the OPL.