Glaucoma is one of the leading causes of irreversible vision loss. Many approaches have recently been proposed for automatic glaucoma detection based on fundus images. However, none of the existing approaches can efficiently remove high redundancy in fundus images for glaucoma detection, which may reduce the reliability and accuracy of glaucoma detection. To avoid this disadvantage, this paper proposes an attention-based convolutional neural network (CNN) for glaucoma detection, called AG-CNN. Specifically, we first establish a large-scale attention-based glaucoma (LAG) database, which includes 11 760 fundus images labeled as either positive glaucoma (4878) or negative glaucoma (6882). Among the 11 760 fundus images, the attention maps of 5824 images are further obtained from ophthalmologists through a simulated eye-tracking experiment. Then, a new structure of AG-CNN is designed, including an attention prediction subnet, a pathological area localization subnet, and a glaucoma classification subnet. The attention maps are predicted in the attention prediction subnet to highlight the salient regions for glaucoma detection, under a weakly supervised training manner. In contrast to other attention-based CNN methods, the features are also visualized as the localized pathological area, which are further added in our AG-CNN structure to enhance the glaucoma detection performance. Finally, the experiment results from testing over our LAG database and another public glaucoma database show that the proposed AG-CNN approach significantly advances the state-of-the-art in glaucoma detection.

A Large-Scale Database and a CNN Model for Attention-Based Glaucoma Detection

Intra-retinal layer segmentation of 3D optical coherence tomography using coarse grained diffusion map

Optical coherence tomography (OCT) is a recently established imaging technique to describe different information about the internal structures of an object and to image various aspects of biological tissues. OCT image segmentation is mostly introduced on retinal OCT to localize the intra-retinal boundaries. Here, we review some of the important image segmentation methods for processing retinal OCT images. We may classify the OCT segmentation approaches into five distinct groups according to the image domain subjected to the segmentation algorithm. Current researches in OCT segmentation are mostly based on improving the accuracy and precision, and on reducing the required processing time. There is no doubt that current 3-D imaging modalities are now moving the research projects toward volume segmentation along with 3-D rendering and visualization. It is also important to develop robust methods capable of dealing with pathologic cases in OCT imaging.

/pdf/a-review-of-algorithms-for-segmentation-of-optical-coherence-gikypa9rec.pdf

A review of algorithms for segmentation of optical coherence tomography from retina.

Optical coherence tomography (OCT) is a powerful and noninvasive method for retinal imaging. In this paper, we introduce a fast segmentation method based on a new variant of spectral graph theory named diffusion maps. The research is performed on spectral domain (SD) OCT images depicting macular and optic nerve head appearance. The presented approach does not require edge-based image information and relies on regional image texture. Consequently, the proposed method demonstrates robustness in situations of low image contrast or poor layer-to-layer image gradients. Diffusion mapping is applied to 2D and 3D OCT datasets composed of two steps, one for partitioning the data into important and less important sections, and another one for localization of internal this http URL the first step, the pixels/voxels are grouped in rectangular/cubic sets to form a graph node.The weights of a graph are calculated based on geometric distances between pixels/voxels and differences of their mean intensity.The first diffusion map clusters the data into three parts, the second of which is the area of interest. The other two sections are eliminated from the remaining calculations. In the second step, the remaining area is subjected to another diffusion map assessment and the internal layers are localized based on their textural similarities.The proposed method was tested on 23 datasets from two patient groups (glaucoma and normals). The mean unsigned border positioning errors(mean - SD) was 8.52 - 3.13 and 7.56 - 2.95 micrometer for the 2D and 3D methods, respectively.

Intra-Retinal Layer Segmentation of 3D Optical Coherence Tomography Using Coarse Grained Diffusion Map

Most current systems for automated glaucoma detection in fundus images rely on segmentation-based features, which are known to be influenced by the underlying segmentation methods. Convolutional Neural Networks (CNNs) are powerful tools for solving image classification tasks as they are able to learn highly discriminative features from raw pixel intensities. However, their applicability to medical image analysis is limited by the non-availability of large sets of annotated data required for training. In this article we present results of analysis of the viability of using CNNs that are pre-trained from non-medical data for automated glaucoma detection. Two different CNNs, namely OverFeat and VGG-S, were applied to fundus images to generate feature vectors. Preprocessing techniques such as vessel inpainting, contrast-limited adaptive histogram equalization (CLAHE) or cropping around the optic nerve head (ONH) area were explored within this framework to evaluate the improvement in feature discrimination, combined with both e1 and e2 regularized logistic regression models. Results on the Drishti-GS1 dataset, evaluated in terms of area under the average ROC curve, suggests the viability of this approach and offer significant evidence of the importance of well-chosen image pre-processing for transfer learning when the amount of data is not sufficient for fine-tuning the network.

/pdf/convolutional-neural-network-transfer-for-automated-glaucoma-2ntfy46z19.pdf

Convolutional neural network transfer for automated glaucoma identification

Thickness related textural properties of retinal nerve fiber layer in color fundus images.

To assess the combined diagnostic power of frequency-doubling technique (FDT)-perimetry and retinal nerve fibre layer (RNFL) thickness measurements with spectral domain optical coherence tomography (SDOCT) The study included 330 experienced participants in five age-related groups: 77 ‘preperimetric’ open-angle glaucoma (OAG) patients, 52 ‘early’ OAG, 50 ‘moderate’ OAG, 54 ocular hypertensivepatients, and 97 healthy subjects For glaucoma assessment in all subjects conventional perimetry, evaluation of fundus photographs, FDT-perimetry and RNFL thickness measurement with SDOCT was done Glaucomatous visual field defects were classified using the Glaucoma Staging System FDT evaluation used a published method with casewise calculation of an ‘FDT-score’, including all missed localized probability levels SDOCT evaluation used mean RNFL thickness and a new individual SDOCT-score considering normal confidence limits in 32 sectors of a peripapillary circular scan To examine the joined value of both methods a combined score was introduced Significance of the difference between Receiver-operating-characteristic (ROC) curves was calculated for a specificity of 96% Sensitivity in the preperimetric glaucoma group was 44% for SDOCT-score, 25% for FDT-score, and 44% for combined score, in the early glaucoma group 83, 81, and 89%, respectively, and in the moderate glaucoma group 94, 94, and 98%, respectively, all at a specificity of 96% ROC performance of the newly developed combined score is significantly above single ROC curves of FDT-score in preperimetric and early OAG and above RNFL thickness in moderate OAG Combination of function and morphology by using the FDT-score and the SDOCT-score performs equal or even better than each single method alone

/pdf/frequency-doubling-technique-perimetry-and-spectral-domain-y8p6rd502g.pdf

Frequency doubling technique perimetry and spectral domain optical coherence tomography in patients with early glaucoma

Automatic Nerve Fiber Layer Segmentation and Geometry Correction on Spectral Domain OCT Images Using Fuzzy C-Means Clustering

This paper proposes a novel super-resolution framework to reconstruct high-resolution fundus images from multiple low-resolution video frames in retinal fundus imaging. Natural eye movements during an examination are used as a cue for super-resolution in a robust maximum a-posteriori scheme. In order to compensate heterogeneous illumination on the fundus, we integrate retrospective illumination correction for photometric registration to the underlying imaging model. Our method utilizes quality self-assessment to provide objective quality scores for reconstructed images as well as to select regularization parameters automatically. In our evaluation on real data acquired from six human subjects with a low-cost video camera, the proposed method achieved considerable enhancements of low-resolution frames and improved noise and sharpness characteristics by 74%. In terms of image analysis, we demonstrate the importance of our method for the improvement of automatic blood vessel segmentation as an example application, where the sensitivity was increased by 13% using super-resolution reconstruction.

/pdf/multi-frame-super-resolution-with-quality-self-assessment-2xsr6jv1f1.pdf

Multi-frame super-resolution with quality self-assessment for retinal fundus videos.

We describe a low-cost, easy to use binocular instrument to acquire retinal video sequences of both eyes simultaneously. After image registration, cardiac cycle-induced pulsatile light attenuation changes can be measured quantitatively with high spatial and temporal resolution. Parameters such as amplitude, pulse form, and time shift between light attenuation changes can be calculated and compared between eye sides. Deviation from inter-eye symmetry can be not only an early sign of beginning eye diseases such as glaucoma but also a sign of pathological changes in the carotid arteries; hence, this method can improve the early detection of pathological changes. Important features compared to existing monocular instruments are a narrow band light source with the wavelength close to the peak of the blood extinction, and a proportional relationship of image intensity and light intensity, which are the main requirements for quantitative evaluation.

R. P. Tornow

Papers

Thickness related textural properties of retinal nerve fiber layer in color fundus images.

Frequency doubling technique perimetry and spectral domain optical coherence tomography in patients with early glaucoma

Automatic Nerve Fiber Layer Segmentation and Geometry Correction on Spectral Domain OCT Images Using Fuzzy C-Means Clustering

Multi-frame super-resolution with quality self-assessment for retinal fundus videos.

Time-resolved quantitative inter-eye comparison of cardiac cycle-induced blood volume changes in the human retina.