A CNN-Based Framework for Automatic Vitreous Segemntation from OCT Images

Hagagg, S.; Khalifa, Fahmi; Abdeltawab, Hisham; Elnakib, Ahmed; Abdelazim, Mohamed; Ghazal, Mohammed; Sandhu, Harpal S.; El-Baz, Ayman

doi:10.1109/ist48021.2019.9010133

Cited by 4 publications

(13 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The obtained p-values (shown in Table 1) illustrate that there is a statistically significant difference (p-value ≤ 0.05) of the two methods. Sample of segmented images of our proposed approach (first row) compared with previous results using only U-Net [16] (second row). The green and yellow colors represents the ground truth and the CNN-segmented respectively.…”

Section: Overall Segmentation Evaluationmentioning

confidence: 99%

“…In contrast to [16], where the OCT images were directly applied to train the U-CNN, the first stage of the proposed CAD system trains the U-CNN model using a fused image (FI) dataset, which integrates the information of the original image with a proposed distance map, and a proposed adaptive appearance map (AAP), instead of the direct original images. • Compared to previous work, the first stage of the proposed CAD system shows superior performance in vitreous segmentation from the OCT images in spite of the great similarity between the vitreous and the background.…”

Section: Introductionmentioning

confidence: 99%

“…Figure 6. Sample of segmented images of our proposed approach (first row) compared with previous results using only U-Net[16] (second row). The green and yellow colors represents the ground truth and the CNN-segmented respectively.…”

mentioning

confidence: 99%

“…Segmentation accuracy comparison of the proposed approach and the previous technique based on U-CNN only[16], using both area and distance-based metrics.…”

mentioning

confidence: 99%

See 3 more Smart Citations

An Automated CAD System for Accurate Grading of Uveitis Using Optical Coherence Tomography Images

Haggag

Khalifa

Abdeltawab

et al. 2021

Sensors

View full text Add to dashboard Cite

Uveitis is one of the leading causes of severe vision loss that can lead to blindness worldwide. Clinical records show that early and accurate detection of vitreous inflammation can potentially reduce the blindness rate. In this paper, a novel framework is proposed for automatic quantification of the vitreous on optical coherence tomography (OCT) with particular application for use in the grading of vitreous inflammation. The proposed pipeline consists of two stages, vitreous region segmentation followed by a neural network classifier. In the first stage, the vitreous region is automatically segmented using a U-net convolutional neural network (U-CNN). For the input of U-CNN, we utilized three novel image descriptors to account for the visual appearance similarity of the vitreous region and other tissues. Namely, we developed an adaptive appearance-based approach that utilizes a prior shape information, which consisted of a labeled dataset of the manually segmented images. This image descriptor is adaptively updated during segmentation and is integrated with the original greyscale image and a distance map image descriptor to construct an input fused image for the U-net segmentation stage. In the second stage, a fully connected neural network (FCNN) is proposed as a classifier to assess the vitreous inflammation severity. To achieve this task, a novel discriminatory feature of the segmented vitreous region is extracted. Namely, the signal intensities of the vitreous are represented by a cumulative distribution function (CDF). The constructed CDFs are then used to train and test the FCNN classifier for grading (grade from 0 to 3). The performance of the proposed pipeline is evaluated on a dataset of 200 OCT images. Our segmentation approach documented a higher performance than related methods, as evidenced by the Dice coefficient of 0.988 ± 0.01 and Hausdorff distance of 0.0003 mm ± 0.001 mm. On the other hand, the FCNN classification is evidenced by its average accuracy of 86%, which supports the benefits of the proposed pipeline as an aid for early and objective diagnosis of uvea inflammation.

show abstract

Section: Overall Segmentation Evaluationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

“…Segmentation accuracy comparison of the proposed approach and the previous technique based on U-CNN only[16], using both area and distance-based metrics.…”

mentioning

confidence: 99%

See 2 more Smart Citations

An Automated CAD System for Accurate Grading of Uveitis Using Optical Coherence Tomography Images

Haggag

Khalifa

Abdeltawab

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…With the development of IR imaging and computer vision [1], image segmentation which aims to extract object of interest from image plays an essential Minjie Wan E-mail: minjiewan1992@njust.edu.cn Guohua Gu E-mail: gghjust@mail.njust.edu.cn 1 School of Electronic and Optical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China 2 Shanghai Institute of Spaceflight Control Technology, Shanghai 201109, China role in many areas of both civil and military applications, such as geology exploration [2], aerospace engineering [3], security monitoring [4] and so on. Among various image segmentation methods [5][6][7][8], ACM has gained popularity because of its excellent ability to obtain closed contours with sub-pixel accuracy [9]. Although, a number of ACMs [10][11][12][13] have achieved satisfactory performances in clear visible images, they only use single image feature information to construct the energy function.…”

Section: Introductionmentioning

confidence: 99%

Multi-Feature Driven Active Contour Segmentation Model for Infrared Image With Intensity Inhomogeneity

Huang¹,

Zhou²,

Wan³

et al. 2021

Preprint

View full text Add to dashboard Cite

Active contour model (ACM) is one of the most widely used image segmentation tools at present, but the existing methods only utilize single feature information of image to minimize the energy function, which is easy to cause false segmentations in infrared (IR) images. In this paper, we propose a multi-feature driven active contour segmentation model to handle IR images with intensity inhomogeneity. Firstly, an especially-designed signed pressure force (SPF) function is constructed by combining the global information calculated by global average gray information and the local multi-feature information calculated by local entropy, local standard deviation and gradient information. Then, we draw upon adaptive weight coefficient computed by local range to incorporate the afore-mentioned global term and local term. Next, the SPF function is substituted into the level set formulation (LSF) for further evolution. Finally, the LSF converges after a finite number of iterations and the IR image segmentation result is obtained from the corresponding convergence result. Experimental results demonstrate that the presented method outperforms typical models in terms of precision rate and overlapping rate in IR test images.

show abstract

Multi-feature driven active contour segmentation model for infrared image with intensity inhomogeneity

et al. 2021

View full text Add to dashboard Cite

A CNN-Based Framework for Automatic Vitreous Segemntation from OCT Images

Cited by 4 publications

References 16 publications

An Automated CAD System for Accurate Grading of Uveitis Using Optical Coherence Tomography Images

An Automated CAD System for Accurate Grading of Uveitis Using Optical Coherence Tomography Images

Multi-Feature Driven Active Contour Segmentation Model for Infrared Image With Intensity Inhomogeneity

Multi-feature driven active contour segmentation model for infrared image with intensity inhomogeneity

Contact Info

Product

Resources

About