Noise reduction in optical coherence tomography images using a deep neural network with perceptually-sensitive loss function.

RETRACTED: Deep learning network (DL-Net) based classification and segmentation of multi-class retinal diseases using OCT scans

As a powerful diagnostic tool, optical coherence tomography (OCT) has been widely used in various clinical setting. However, OCT images are susceptible to inherent speckle noise that may contaminate subtle structure information, due to low-coherence interferometric imaging procedure. Many supervised learning-based models have achieved impressive performance in reducing speckle noise of OCT images trained with a large number of noisy-clean paired OCT images, which are not commonly feasible in clinical practice. In this article, we conducted a comparative study to investigate the denoising performance of OCT images over different deep neural networks through an unsupervised Noise2Noise (N2N) strategy, which only trained with noisy OCT samples. Four representative network architectures including U-shaped model, multi-information stream model, straight-information stream model and GAN-based model were investigated on an OCT image dataset acquired from healthy human eyes. The results demonstrated all four unsupervised N2N models offered denoised OCT images with a performance comparable with that of supervised learning models, illustrating the effectiveness of unsupervised N2N models in denoising OCT images. Furthermore, U-shaped models and GAN-based models using UNet network as generator are two preferred and suitable architectures for reducing speckle noise of OCT images and preserving fine structure information of retinal layers under unsupervised N2N circumstances.

Age-related macular degeneration (ARMD) is a degenerative eye disease among people over the age of 50. This disorder affects the retina and macula of the eye. The macula is the central part of the retina. The function of the macular region is to give a sharp and clear vision. ARMD is mainly identified by the presence of yellow deposits called drusen and the structural changes in the retinal pigment epithelium (RPE) layer of the retina. If ARMD is not identified at an early stage, it will lead to permanent blindness. The ARMD is mainly diagnosed using Color fundus and Optical coherence tomography (OCT) images of the retina. OCT is an imaging technique that makes use of low-coherence light to capture two-and three-dimensional images from within optical scattering media. OCT images usually are at a micrometer resolution, affected by a noise called speckle noise. If speckle noise is present in an OCT image, diagnosis of ARMD is very difficult, irrespective of whether it is manually or automatically. So, removal of speckle-noise is significant for the accurate diagnosis and early detection of ARMD. In this paper, we evaluate the performance of eight popular image noise reduction filters for its ability to remove speckle noise in terms of different performance measures on a publicly available OCT dataset. Our evaluation shows that, out of the filters selected for assessment, a fast non-local mean filter outperforms all other filters regarding all the performance measures used.

Aim To develop an optimal method for automated segmentation of atherosclerotic plaque structural components in optical coherence tomography (OCT) images using an ensemble of deep learning neural network models based on a comparison of nine artificial neural network architectures.Material and methods This study utilized a multidisciplinary OCT dataset obtained from examinations of 103 patients. Each OCT session was annotated as a set of two-dimensional binary masks corresponding to the pixel boundaries of four key morphological features of plaques: vascular lumen, fibrous cap, lipid core, and microvessels. Nine deep learning models, including U-Net, DeepLabV3, and others, were used to segment the anatomical structures. Model hyperparameters were tuned using Bayesian optimization, and the performance was assessed using the Dice Similarity Coefficient (DSC) metric.Results The models demonstrated high segmentation accuracy for the vascular lumen (DSC: 0.987) and moderate accuracy for the fibrous cap (DSC: 0.736) and lipid core (DSC: 0.751). Microvessel segmentation proved more challenging, with a final DSC accuracy of 61%. A weighted ensemble of models, accounting for the uneven representation of different morphological structures and model confidence, achieved an average DSC of 88.2%, which significantly improved the overall accuracy compared to individual models. This accuracy value exceeds those of all individual models, including the highest DSC values of 0.784 (for microvessels) and 0.751 (for lipid core), indicating a consistent improvement in the segmentation accuracy by integrating the models.Conclusion The study confirms the effectiveness of the proposed strategy for segmentation of atherosclerotic plaques in OCT images. This strategy is based on using specialized models for various morphological features and a weighted ensemble adapted to the uneven representation of different morphological structures and morphological complexity. This approach provided a high segmentation accuracy despite the pronounced inequality in the representation of classes. The study results can contribute to the development of decision support methods in cardiology aimed at improving diagnostic accuracy and treatment of cardiovascular diseases.

Microaneurysms (MAs) are pathognomonic signs that help clinicians to detect diabetic retinopathy (DR) in the early stages. Automatic detection of MA in retinal images is an active area of research due to its application in screening processes for DR which is one of the main reasons of blindness amongst the working-age population. The focus of these works is on the automatic detection of MAs in en face retinal images like fundus color and Fluorescein Angiography (FA). On the other hand, detection of MAs from Optical Coherence Tomography (OCT) images has 2 main advantages: first, OCT is a non-invasive imaging technique that does not require injection, therefore is safer. Secondly, because of the proven application of OCT in detection of Age-Related Macular Degeneration, Diabetic Macular Edema, and normal cases, thanks to detecting MAs in OCT, extensive information is obtained by using this imaging technique. In this research, the concentration is on the diagnosis of MAs using deep learning in the OCT images which represent in-depth structure of retinal layers. To this end, OCT B-scans should be divided into strips and MA patterns should be searched in the resulted strips. Since we need a dataset comprising OCT image strips with suitable labels and such large labelled datasets are not yet available, we have created it. For this purpose, an exact registration method is utilized to align OCT images with FA photographs. Then, with the help of corresponding FA images, OCT image strips are created from OCT B-scans in four labels, namely MA, normal, abnormal, and vessel. Once the dataset of image strips is prepared, a stacked generalization (stacking) ensemble of four fine-tuned, pre-trained convolutional neural networks is trained to classify the strips of OCT images into the mentioned classes. FA images are used once to create OCT strips for training process and they are no longer needed for subsequent steps. Once the stacking ensemble model is obtained, it will be used to classify the OCT strips in the test process. The results demonstrate that the proposed framework classifies overall OCT image strips and OCT strips containing MAs with accuracy scores of 0.982 and 0.987, respectively.

Objective. The quality of optical coherence tomography (OCT) en face images is crucial for clinical visualization of early disease. As a three dimensional and coherent imaging, defocus and speckle noise are inevitable, which seriously affect evaluation of microstructure of bio-samples in OCT images. The deep learning has demonstrated great potential in OCT refocusing and denoising, but it is limited by the difficulty of sufficient paired training data. This work aims to develop an unsupervised method to enhance the quality of OCT en face images. Approach. We proposed an unsupervised deep learning-based pipeline. The unregistered defocused conventional OCT images and focused speckle-free OCT images were collected by a home-made speckle modulating OCT system to construct the dataset. The image enhancement model was trained with the cycle training strategy. Finally, the speckle noise and defocus were both effectively improved. Main results. The experimental results on complex bio-samples indicated that the proposed method is effective and generalized in enhancing the quality of OCT en face images. Significance. The proposed unsupervised deep learning method helps to reduce the complexity of data construction, which is conducive to practical applications in OCT bio-sample imaging.

As a new non-invasive and high-resolution scanning method, optical coherence tomography (OCT) has been widely used in clinical practice, but OCT images have serious speckle noise, which greatly affects the diagnosis of diseases. Two original dictionary noise reduction algorithms have been improved for multiplicative speckle noise in OCT. The algorithm first performs logarithmic transformation on OCT images, uses orthogonal matching pursuit algorithm for sparse coding, and K singular value decomposition learning algorithm to update adaptive dictionary. Finally, it returns to the space domain through weighted average and exponential transformation. The experimental results show that the improved two dictionary algorithms can effectively reduce the speckle noise in OCT images and obtain good visual effects. The noise reduction effect is evaluated by four factors: mean squared error (MSE), peak signal-to-noise ratio (PSNR), structural similarity (SSIM) and edge-preserving index (EPI). Compared with the two original dictionary noise reduction algorithms and the traditional filtering algorithms, the noise reduction effect of the two improved dictionary algorithms is better than that of other algorithms, and the improved adaptive dictionary algorithm performs better.

Previous studies have demonstrated that optical coherence tomography (OCT) could be used to delineate alterations in the microstructure of cartilage, and have suggested that changes in the polarization state of light as detected by OCT could provide information on the birefringence properties of articular cartilage as influenced by disease. In this study we have used both OCT and polarization sensitive optical coherence tomography (PS-OCT) technologies to evaluate normal and abnormal bovine articular cartilage according to established structural, organizational, and birefringent characteristics of degenerative joint disease (DJD) in order to determine if this technology can be used to differentiate various stages of DJD as a minimally invasive imaging tool. Fresh bovine femoral-tibial joints were obtained from an abattoir, and 45 cartilage specimens were harvested from 8 tibial plateaus. Whole ex vivo specimens of normal and degenerative articular cartilage were imaged by both OCT and PS-OCT, then fixed and processed for histological evaluation. OCT/PS-OCT images and corresponding histology sections of each specimen were scored according to a modified Mankin structural grading scale and compared. OCT and PS-OCT imaging allowed structural evaluation of intact articular cartilage along a 6 mm surface length to a depth of 2 mm with a transverse resolution of 12 microm and an axial resolution of 10 microm. The OCT and PS-OCT images demonstrated characteristic alterations in the structure of articular cartilage with a high correlation to histological evaluation (kappa = 0.776). The OCT images were able to demonstrate early to advanced structural changes of articular cartilage while the optical phase retardation images obtained by PS-OCT imaging were able to discriminate areas where disorganization of the cartilage matrix was present, however, these characteristics are much different than those reported where OCT images alone were used to characterize tissue birefringence. No evidence of differences in OCT or PS-OCT images were detected between specimens of similar structural characteristics where proteoglycan was judged present or absent by safranin-O Fast Green staining. The combined use of OCT and PS-OCT technologies to obtain images from a single system is able to demonstrate and discriminate between characteristics of very early stages of surface irregularities not previously reported for OCT imaging, to deep clefts and collagen matrix disorganization for tissue at depths of up to 2 mm with good correlation to histology. PS-OCT and accumulated optical phase retardation images of articular cartilage as constructed from alterations in Stokes vector parameters appear to give a valuable but different assessment of alterations in tissue birefringence and organization than have been reported for OCT images obtained with the use of polarized or non-polarized light sources. This is the first time that alterations in the polarization state of light reflected from within the tissue have been demonstrated to be consistent with changes observed in the orientation and organization of the collagen matrix in advanced stages of DJD. The degree of phase transformation of light reflected from within the tissue as determined by PS-OCT imaging does not appear to be altered by the presence or absence of proteoglycan.

Deep neural network (DNN) can extract high-dimensional feature of images for computer vision tasks including Optical Coherence Tomography (OCT) images classification. However, OCT images are usually processed by DNN just like natural images, thus the performance of DNN is not satisfactory. We present an end-to-end DNN targeting OCT images classification. Considering the characteristic of OCT images, we introduce attention mechanism into classifier to extract more specific feature of OCT images. Our network demonstrates its capacity to enhance the features that represent the disease region. Our method achieves the state-of-the-art performance with average accuracy of 99.5% and F1-score of 0.995 on the OCT images dataset.

Optical coherence tomography (OCT) is becoming one of the most important detection modalities for fast and noninvasive assessment of ophthalmological diseases. Diabetic macular edema (DME) is one of the important reasons leads to blindness. Its pathological features are mainly manifested in the accumulation of fluid in the retina. An automated method is proposed to identify and quantify the volume of cystoid macular edema in Spectral Domain OCT (SD-OCT) images. In the first stage of preprocessing, we balance the apparent signal-to-noise of each retinal OCT image. Because the signal-to-noise of OCT images is variable from patient to patient, and balance of the signal-to-noise ensures consistent segmentation of cystoid fluid. Speckle noise is the main reason leads to quality degrading in OCT images. The denoising method should be efficient for the noise suppression, and the edge information can be preserved at the same time. Then we used the anisotropic diffusion filter to suppress shot noise. The intensity inhomogeneity in OCT images may lead to false detection in the further segmentation work. Then we used the gamma transformation to change the brightness, which eliminates the effect availably. In the second stage of segmentation, we solve the problem of segmentation effectively by the improved level set method and calculated the area of edema area, which provides quantitative analytic tools for clinical diagnosis and therapy. Finally, the proposed method was evaluated on 15 SD-OCT retinal images from DME adults. Leave-one-out evaluation resulted in a precision, sensitivity and dice similarity coefficient (DSC) of 81.12%, 86.90% and 80.05%, respectively.

Retinal layer thickness, evaluated as a function of spatial position from optical coherence tomography (OCT) images is an important diagnostics marker for many retinal diseases. However, due to factors such as speckle noise, low image contrast, irregularly shaped morphological features such as retinal detachments, macular holes, and drusen, accurate segmentation of individual retinal layers is difficult. To address this issue, a computer method for retinal layer segmentation from OCT images is presented. An efficient two-step kernel-based optimization scheme is employed to first identify the approximate locations of the individual layers, which are then refined to obtain accurate segmentation results for the individual layers. The performance of the algorithm was tested on a set of retinal images acquired in-vivo from healthy and diseased rodent models with a high speed, high resolution OCT system. Experimental results show that the proposed approach provides accurate segmentation for OCT images affected by speckle noise, even in sub-optimal conditions of low image contrast and presence of irregularly shaped structural features in the OCT images.

The purpose of this study was to develop and evaluate the performance of a convolutional neural network (CNN) that uses a novel A-line based classification approach to detect cancer in OCT images of breast specimens. Deep learning algorithms have been developed for OCT ophthalmology applications using pixel-based classification approaches. In this study, a novel deep learning approach was developed that classifies OCT A-lines of breast tissue. De-identified human breast tissues from mastectomy and breast reduction specimens were excised from patients at Columbia University Medical Center. A total of 82 specimens from 49 patients were imaged with OCT, including both normal tissues and non-neoplastic tissues. The proposed algorithm utilized a hybrid 2D/1D convolutional neural network (CNN) to map each single B-scan to a 1D label vector, which were derived from manual annotation. Each A-line was labelled as one of the following tissue types: ductal carcinoma in situ (DCIS), invasive ductal carcinoma (IDC), adipose, and stroma. Five-fold cross-validation Dice scores across tissue types were: 0.82-0.95 for IDC, 0.54-0.75 for DCIS, 0.67-0.91 for adipose, and 0.61-0.86 for stroma. In a second experiment, IDC and DCIS were combined as a single tissue class (malignancy) while stroma and adipose were combined as a second tissue class (non-malignancy). In this setup, the experiment yielded five-fold cross-validation Dice scores between 0.89-0.93, respectively. Future work includes acquiring more patient samples and to compare the algorithm to previous works, including both deep learning and traditional automatic image processing methods for classification of breast tissue in OCT images.

Collaborative masking based speckle disentanglement for self-supervised optical coherence tomography image despeckling

In this paper, we propose a total variation based on block matching 3D (BM3D-TV method), which includes the total variation regular term, the data fidelity term, and the block matching term. In addition, we also propose a fast numerical algorithm based on the split Bregman iteration for the proposed method. By assigning suitable weights to the data fidelity term and block matching term, the image noise reduction and the image structural characteristics can be matched optimally. We test the proposed method on six human retinal and one mouse skin optical coherence tomography (OCT) images respectively, and also compare it with total variation (TV) and BM3D, which were proved to be effective in denoising. The performances of these methods are quantitatively evaluated in terms of the signal-to-noise ratio, the contrast-to-noise ratio, and the averaged equivalent number of homogeneous areas at the aspects of speckle reduction and structure protection. Vast experiments show that the BM3D-TV method can effectively reduce speckle noise in OCT images, protect important structural information and improve image quality, compared with the BM3D and TV methods.

High-Definition and 3-dimensional Imaging of Macular Pathologies with High-speed Ultrahigh-Resolution Optical Coherence Tomography