Search

Background: Pancreatic ductal adenocarcinoma (PDAC) is the most prevalent type of pancreas cancer with a high mortality rate and its staging is highly dependent on the extent of involvement between the tumor and surrounding vessels, facilitating treatment response assessment in PDAC. Objective: This study aims at detecting and visualizing the tumor region and the surrounding vessels in PDAC CT scan since, despite the tumors in other abdominal organs, clear detection of PDAC is highly difficult. Material and Methods: This retrospective study consists of three stages: 1) a patch-based algorithm for differentiation between tumor region and healthy tissue using multi-scale texture analysis along with L1-SVM (Support Vector Machine) classifier, 2) a voting-based approach, developed on a standard logistic function, to mitigate false detections, and 3) 3D visualization of the tumor and the surrounding vessels using ITK-SNAP software. Results: The results demonstrate that multi-scale texture analysis strikes a balance between recall and precision in tumor and healthy tissue differentiation with an overall accuracy of 0.78±0.12 and a sensitivity of 0.90±0.09 in PDAC. Conclusion: Multi-scale texture analysis using statistical and wavelet-based features along with L1-SVM can be employed to differentiate between healthy and pancreatic tissues. Besides, 3D visualization of the tumor region and surrounding vessels can facilitate the assessment of treatment response in PDAC. However, the 3D visualization software must be further developed for integrating with clinical applications. [ABSTRACT FROM AUTHOR]

Background: Asymmetry analysis of retinal layers in right and left eyes can be a valuable tool for early diagnoses of retinal diseases. To determine the limits of the normal interocular asymmetry in retinal layers around macula, thickness measurements are obtained with optical coherence tomography (OCT). Methods: For this purpose, after segmentation of intraretinal layer in threedimensional OCT data and calculating the midmacular point, the TM of each layer is obtained in 9 sectors in concentric circles around the macula. To compare corresponding sectors in the right and left eyes, the TMs of the left and right images are registered by alignment of retinal raphe (i.e. diskfovea axes). Since the retinal raphe of macular OCTs is not calculable due to limited region size, the TMs are registered by first aligning corresponding retinal raphe of fundus images and then registration of the OCTs to aligned fundus images. To analyze the asymmetry in each retinal layer, the mean and standard deviation of thickness in 9 sectors of 11 layers are calculated in 50 normal individuals. Results: The results demonstrate that some sectors of retinal layers have signifcant asymmetry with P < 0.05 in normal population. In this base, the tolerance limits for normal individuals are calculated. Conclusion: This article shows that normal population does not have identical retinal information in both eyes, and without considering this reality, normal asymmetry in information gathered from both eyes might be interpreted as retinal disorders. [ABSTRACT FROM AUTHOR]

Nowadays, it is obvious that there is a relationship between changes in the retinal vessel structure and diseases such as diabetic, hypertension, stroke, and the other cardiovascular diseases in adults as well as retinopathy of prematurity in infants. Retinal fundus images provide non-invasive visualization of the retinal vessel structure. Applying image processing techniques in the study of digital color fundus photographs and analyzing their vasculature is a reliable approach for early diagnosis of the aforementioned diseases. Reduction in the arteriolar-venular ratio of retina is one of the primary signs of hypertension, diabetic, and cardiovascular diseases which can be calculated by analyzing the fundus images. To achieve a precise measuring of this parameter and meaningful diagnostic results, accurate classification of arteries and veins is necessary. Classification of vessels in fundus images faces with some challenges that make it difficult. In this paper, a comprehensive study of the proposed methods for classification of arteries and veins in fundus images is presented. Considering that these methods are evaluated on different datasets and use different evaluation criteria, it is not possible to conduct a fair comparison of their performance. Therefore, we evaluate the classification methods from modeling perspective. This analysis reveals that most of the proposed approaches have focused on statistics, and geometric models in spatial domain and transform domain models have received less attention. This could suggest the possibility of using transform models, especially data adaptive ones, for modeling of the fundus images in future classification approaches. [ABSTRACT FROM AUTHOR]

Diagnosis of corneal diseases is possible by measuring and evaluation of corneal thickness in different layers. Thus, the need for precise segmentation of corneal layer boundaries is inevitable. Obviously, manual segmentation is time-consuming and imprecise. In this paper, the Gaussian mixture model (GMM) is used for automatic segmentation of three clinically important corneal boundaries on optical coherence tomography (OCT) images. For this purpose, we apply the GMM method in two consequent steps. In the first step, the GMM is applied on the original image to localize the first and the last boundaries. In the next step, gradient response of a contrast enhanced version of the image is fed into another GMM algorithm to obtain a more clear result around the second boundary. Finally, the first boundary is traced toward down to localize the exact location of the second boundary. We tested the performance of the algorithm on images taken from a Heidelberg OCT imaging system. To evaluate our approach, the automatic boundary results are compared with the boundaries that have been segmented manually by two corneal specialists. The quantitative results show that the proposed method segments the desired boundaries with a great accuracy. Unsigned mean errors between the results of the proposed method and the manual segmentation are 0.332, 0.421, and 0.795 for detection of epithelium, Bowman, and endothelium boundaries, respectively. Unsigned mean errors of the inter-observer between two corneal specialists have also a comparable unsigned value of 0.330, 0.398, and 0.534, respectively. [ABSTRACT FROM AUTHOR]

Background: Early diagnosis of diseases is an important step in their treatment process, but it is difficult due to lack of satisfactorily sensitive imaging devices and difference in the parameters of the subjects. Asymmetry analysis of retinal nerve fiber layer thickness and total retinal thickness can provide a criterion for physicians to accomplish early diagnosis. Methods: Data were collected from 50 normal subjects (aged 35 ± 10 years) by a Topcon model of 3D-OCT 1000. For this purpose, a pipeline of procedures was utilized; segmentation of retinal layers by diffusion map method, scaling operator, automatic localization of the center of macula, and calculation of mean and SD. Findings: The maximum of average RNFL and total retina thickness in right and left eyes is seen in the perifoveal nasal, and the minimum is seen in the fovea. The results demonstrate an overall symmetry between the two eyes (P > 0.05). Tolerance limits in RNFL of normal subjects for 9 sectors (from 1 to 9), respectively, are 1.16 ± 0.99, 2.43 ± 1.57, 3.96 ± 2.08, 1.11 ± 1.09, 2.83 ± 1.66, 3.09 ± 1.39, 4.91 ± 1.79, 0.82 ± 1.25, and 3.31 ± 1.63 μm. Tolerance limits in total retina of normal subjects for 9 sectors, respectively, are 9.84 ± 2.75, 3.60 ± 2.74, 4.82 ± 3.23, 3.33 ± 1.76, 6.51 ± 3.69, 3.23 ± 1.30, 5.96 ± 2.62, 2.79 ± 1.42, 6.72 ± 2.57 μm. Conclusion: Asymmetry analysis can be a criterion for early detection of diseases. In this study asymmetry analysis and tolerance limits were calculated. [ABSTRACT FROM AUTHOR]

Background: Enhanced depth imaging optical coherence tomography images (EDI-OCT) is used for detailed imaging of the choroid layer that contains the highest amount of blood flow in the eye and is affected in several diseases such as choroidal polyps, age-related degeneration and central serous chorioretinopathy. Choroidal segmentation is really important, but the manual segmentation is time consuming and encounters difficulties when large numbers of data is available. Since a large amount of information is available in the images, non-automated and visual analysis of data is almost impossible for the ophthalmologist. The main goal of automatic segmentation was to help the ophthalmologists in the diagnosis and monitoring diseases related to the eye. Methods: The data used in this project was obtained from the Heidelberg OCT-HRA2-KT instrument. Fifty 2 dimensional data were used to evaluate the algorithm. In this study, the retinal pigment epithelium (RPE) and choroid was segmented using a boundary detection algorithm named dynamic programming. Findings: The proposed algorithm was compared with the manual segmentation and the results showed an unsigned error of 1.71 ± 0.93 pixels for retinal pigmented epithelium (RPE) extraction and 10.48 ± 4.11 pixels for choroid detection. It showed significant improvements over other approaches like k-means method. Conclusion: A few automated methods are applied in the choroid segmentation and most of the studies were mainly focused on the manual separation. In this study, a fast and automated method was provided for the segmentation of choroid area. [ABSTRACT FROM AUTHOR]

Purpose: Several machine learning studies have used optical coherence tomography (OCT) for multiple sclerosis (MS) classification with promising outcomes. Infrared reflectance scanning laser ophthalmoscopy (IR-SLO) captures high-resolution fundus images, commonly combined with OCT for fixed B-scan positions. However, no machine learning research has utilized IR-SLO images for automated MS diagnosis., Methods: This study utilized a dataset comprised of IR-SLO images and OCT data from Isfahan, Iran, encompassing 32 MS and 70 healthy individuals. A number of convolutional neural networks (CNNs)-namely, VGG-16, VGG-19, ResNet-50, ResNet-101, and a custom architecture-were trained with both IR-SLO images and OCT thickness maps as two separate input datasets. The highest performing models for each modality were then integrated to create a bimodal model that receives the combination of OCT thickness maps and IR-SLO images. Subject-wise data splitting was employed to prevent data leakage among training, validation, and testing sets., Results: Overall, images of the 102 patients from the internal dataset were divided into test, validation, and training subsets. Subsequently, we employed a bootstrapping approach on the training data through iterative sampling with replacement. The performance of the proposed bimodal model was evaluated on the internal test dataset, demonstrating an accuracy of 92.40% ± 4.1% (95% confidence interval [CI], 83.61-98.08), sensitivity of 95.43% ± 5.75% (95% CI, 83.71-100.0), specificity of 92.82% ± 3.72% (95% CI, 81.15-96.77), area under the receiver operating characteristic (AUROC) curve of 96.99% ± 2.99% (95% CI, 86.11-99.78), and area under the precision-recall curve (AUPRC) of 97.27% ± 2.94% (95% CI, 86.83-99.83). Furthermore, to assess the model generalization ability, we examined its performance on an external test dataset following the same bootstrap methodology, achieving promising results, with accuracy of 85.43% ± 0.08% (95% CI, 71.43-100.0), sensitivity of 97.33% ± 0.06% (95% CI, 83.33-100.0), specificity of 84.6% ± 0.10% (95% CI, 71.43-100.0), AUROC curve of 99.67% ± 0.02% (95% CI, 95.63-100.0), and AUPRC of 99.65% ± 0.02% (95% CI, 94.90-100.0)., Conclusions: Incorporating both modalities improves the performance of automated diagnosis of MS, showcasing the potential of utilizing IR-SLO as a complementary tool alongside OCT., Translational Relevance: Should the results of our proposed bimodal model be validated in future work with larger and more diverse datasets, diagnosis of MS based on both OCT and IR-SLO can be reliably integrated into routine clinical practice.

Optical Coherence Tomography (OCT) is one of the well-known imaging systems in ophthalmology that provides images with high resolution from retinal tissue. However, like other coherent imaging systems, OCT images suffer from speckle noise which decreases the image quality. Denoising can be considered as an estimation problem in a Bayesian framework. So, finding a suitable distribution for noiseless data is an important issue. We propose a statistical model for OCT data, namely Asymmetric Normal Laplace Mixture Model (ANLMM), and then convert its distribution to normal by Gaussianization Transform (GT). Finally, by applying the Spatially Constrained Gaussian Mixture Model (SC-GMM), a new OCT denoising algorithm is introduced, which significantly outperforms the other methods in terms of Contrast-to-Noise Ratio (CNR).

Optical Coherence Tomography (OCT) is known as a non-invasive and high resolution imaging modality in ophthalmology. Effecting noise on the OCT images as well as other reasons cause a random behavior in these images. In this study, we introduce a new statistical model for retinal layers in healthy OCT images. This model, namely asymmetric Normal Laplace (NL), fits well the advent of asymmetry and heavy-tailed in intensity distribution of each layer. Due to the layered structure of retina, a mixture model is addressed. It is proposed to evaluate the fitness criteria called Kull-back Leibler Divergence (KLD) and chi-square test along visual results. The results express the well performance of proposed model in fitness of data except for 6 th and 7 th layers. Using a complicated model, e.g. a mixture model with two component, seems to be appropriate for these layers. The mentioned process for train images can then be devised for a test image by employing the Expectation Maximization (EM) algorithm to estimate the values of parameters in mixture model.