Robust segmentation of intraretinal layers in the normal human fovea using a novel statistical model based on texture and shape analysis

Kajić, Vedran; Považay, Boris; Hermann, B.; Höfer, Bernd; Marshall, David; Rosin, Paul L.; Drexler, Wolfgang

doi:10.1364/oe.18.014730

Cited by 126 publications

(93 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Kajić et al proposed a method that used a large training dataset obtained from manual segmentations by human operators as input to develop a statistical model to segment seven retinal layers. 18 Garvin et al proposed a graph search-based 3D OCT retinal image segmentation algorithm, 19,20 which could segment¯ve retinal layers, which was later extended to incorporate hard/soft constraints. 21 Lee used multi-scale 3D graph search for segmenting the optic nerve head.…”

Section: Introductionmentioning

confidence: 99%

3D automatic segmentation method for retinal optical coherence tomography volume data using boundary surface enhancement

Sun

Zhang

Zhao

et al. 2016

J. Innov. Opt. Health Sci.

View full text Add to dashboard Cite

With the introduction of spectral-domain optical coherence tomography (SD-OCT), much larger image datasets are routinely acquired compared to what was possible using the previous generation of time-domain OCT. Thus, there is a critical need for the development of three-dimensional (3D) segmentation methods for processing these data. We present here a novel 3D automatic segmentation method for retinal OCT volume data. Brie°y, to segment a boundary surface, two OCT volume datasets are obtained by using a 3D smoothing¯lter and a 3D differential¯lter. Their linear combination is then calculated to generate new volume data with an enhanced boundary surface, where pixel intensity, boundary position information, and intensity changes on both sides of the boundary surface are used simultaneously. Next, preliminary discrete boundary points are detected from the A-Scans of the volume data. Finally, surface smoothness constraints and a dynamic threshold are applied to obtain a smoothed boundary surface by correcting a small number of error points. Our method can extract retinal layer boundary surfaces sequentially with a decreasing search region of volume data. We performed automatic segmentation on eight human OCT volume datasets acquired from a commercial Spectralis OCT system, where each volume of datasets contains 97 OCT B-Scan images with a resolution of 496 Â 512 (each B-Scan comprising 512 A-Scans containing 496 pixels); experimental results show that

show abstract

Section: Introductionmentioning

confidence: 99%

3D automatic segmentation method for retinal optical coherence tomography volume data using boundary surface enhancement

Sun

Zhang

Zhao

et al. 2016

J. Innov. Opt. Health Sci.

View full text Add to dashboard Cite

show abstract

“…The standard software of OCT devices includes algorithms for automatic segmentation of layers, which are e.g. based on: analysis of the image brightness [6], the technique of active contours [7], the pattern recognition [8], the graph theory [9][10][11], or techniques of grouping [12,13]. Automatic layer segmentation of the retina in the OCT images requires overcoming many problems, such as noise [14], an uneven reflection of light by the tissues [15], absorption of light through the blood vessels, an unexpected movement of the patient, and the dependence of the proper segmentation algorithm on the device [16].…”

Section: Introductionmentioning

confidence: 99%

Improving Segmentation of 3D Retina Layers Based on Graph Theory Approach for Low Quality OCT Images

Stankiewicz

Marciniak

Dąbrowski

et al. 2016

Metrology and Measurement Systems

View full text Add to dashboard Cite

This paper presents signal processing aspects for automatic segmentation of retinal layers of the human eye. The paper draws attention to the problems that occur during the computer image processing of images obtained with the use of the Spectral Domain Optical Coherence Tomography (SD OCT). Accuracy of the retinal layer segmentation for a set of typical 3D scans with a rather low quality was shown. Some possible ways to improve quality of the final results are pointed out. The experimental studies were performed using the so-called B-scans obtained with the OCT Copernicus HR device.

show abstract

“…Published methods of OCT image analysis have dealt with segmentation of intraretinal layers and tissue structures, and usually rely on certain boundary models or use filtering techniques [7][8][9][10][11]. Besides that, to the best of our knowledge none has so far been presented for industrial applications except for our 2D "ridge detection" method [12].…”

Section: Introductionmentioning

confidence: 99%

Accurate and automated image segmentation of 3D optical coherence tomography data suffering from low signal-to-noise levels

Su¹,

Ekberg²,

Leitner³

et al. 2014

J. Opt. Soc. Am. A

View full text Add to dashboard Cite

Optical coherence tomography (OCT) has proven to be a useful tool for investigating internal structures in ceramic tapes, and the technique is expected to be important for roll-to-roll manufacturing. However, because of high scattering in ceramic materials, noise and speckles deteriorate the image quality, which makes automated quantitative measurements of internal interfaces difficult. To overcome this difficulty we present in this paper an innovative image analysis approach based on volumetric OCT data. The engine in the analysis is a 3D image processing and analysis algorithm. It is dedicated to boundary segmentation and dimensional measurement in volumetric OCT images, and offers high accuracy, efficiency, robustness, subpixel resolution, and a fully automated operation. The method relies on the correlation property of a physical interface and effectively eliminates pixels caused by noise and speckles. The remaining pixels being stored are the ones confirmed to be related to the target interfaces. Segmentation of tilted and curved internal interfaces separated by ∼10 μm in the Z direction is demonstrated. The algorithm also extracts full-field top-view intensity maps of the target interfaces for high-accuracy measurements in the X and Y directions. The methodology developed here may also be adopted in other similar 3D imaging and measurement technologies, e.g., ultrasound imaging, and for various materials.

show abstract

Robust segmentation of intraretinal layers in the normal human fovea using a novel statistical model based on texture and shape analysis

Cited by 126 publications

References 17 publications

3D automatic segmentation method for retinal optical coherence tomography volume data using boundary surface enhancement

3D automatic segmentation method for retinal optical coherence tomography volume data using boundary surface enhancement

Improving Segmentation of 3D Retina Layers Based on Graph Theory Approach for Low Quality OCT Images

Accurate and automated image segmentation of 3D optical coherence tomography data suffering from low signal-to-noise levels

Contact Info

Product

Resources

About