-
1.
Automated Deformation-Based Analysis of 3D Optical Coherence Tomography in Diabetic Retinopathy.
Khansari, MM, Zhang, J, Qiao, Y, Gahm, JK, Sarabi, MS, Kashani, AH, Shi, Y
IEEE transactions on medical imaging. 2020;(1):236-245
-
-
Free full text
-
Abstract
Diabetic retinopathy (DR) is a significant microvascular complication of diabetes mellitus and a leading cause of vision impairment in working age adults. Optical coherence tomography (OCT) is a routinely used clinical tool to observe retinal structural and thickness alterations in DR. Pathological changes that alter the normal anatomy of the retina, such as intraretinal edema, pose great challenges for conventional layer-based analysis of OCT images. We present an alternative approach for the automated analysis of OCT volumes in DR research based on nonlinear registration. In this paper, we first obtain an anatomically consistent volume of interest (VOI) in different OCT images via carefully designed masking and affine registration. After that, efficient B-spline transformations are computed using stochastic gradient descent optimization. Using the OCT volumes of normal controls, for which layer-based segmentation works well, we demonstrate the accuracy of our registration-based analysis in aligning layer boundaries. By nonlinearly registering the OCT volumes of DR subjects to an atlas constructed from normal controls and measuring the Jacobian determinant of the deformation, we can simultaneously visualize tissue contraction and expansion due to DR pathology. Tensor-based morphometry (TBM) can also be performed for quantitative analysis of local structural changes. In our experimental results, we apply our method to a dataset of 105 subjects and demonstrate that volumetric OCT registration and TBM analysis can successfully detect local retinal structural alterations due to DR.
-
2.
Skeletal ciliopathies: a pattern recognition approach.
Handa, A, Voss, U, Hammarsjö, A, Grigelioniene, G, Nishimura, G
Japanese journal of radiology. 2020;(3):193-206
Abstract
Ciliopathy encompasses a diverse group of autosomal recessive genetic disorders caused by mutations in genes coding for components of the primary cilia. Skeletal ciliopathy forms a subset of ciliopathies characterized by distinctive skeletal changes. Common skeletal ciliopathies include Jeune asphyxiating thoracic dysplasia, Ellis-van Creveld syndrome, Sensenbrenner syndrome, and short-rib polydactyly syndromes. These disorders share common clinical and radiological features. The clinical hallmarks comprise thoracic hypoplasia with respiratory failure, body disproportion with a normal trunk length and short limbs, and severely short digits occasionally accompanied by polydactyly. Reflecting the clinical features, the radiological hallmarks consist of a narrow thorax caused by extremely short ribs, normal or only mildly affected spine, shortening of the tubular bones, and severe brachydactyly with or without polydactyly. Other radiological clues include trident ilia/pelvis and cone-shaped epiphysis. Skeletal ciliopathies are commonly associated with extraskeletal anomalies, such as progressive renal degeneration, liver disease, retinopathy, cardiac anomalies, and cerebellar abnormalities. In this article, we discuss the radiological pattern recognition approach to skeletal ciliopathies. We also describe the clinical and genetic features of skeletal ciliopathies that the radiologists should know for them to play an appropriate role in multidisciplinary care and scientific advancement of these complicated disorders.
-
3.
Fundamentals of artificial intelligence for ophthalmologists.
Ahmad, BU, Kim, JE, Rahimy, E
Current opinion in ophthalmology. 2020;(5):303-311
Abstract
PURPOSE OF REVIEW As artificial intelligence continues to develop new applications in ophthalmic image recognition, we provide here an introduction for ophthalmologists and a primer on the mechanisms of deep learning systems. RECENT FINDINGS Deep learning has lent itself to the automated interpretation of various retinal imaging modalities, including fundus photography and optical coherence tomography. Convolutional neural networks (CNN) represent the primary class of deep neural networks applied to these image analyses. These have been configured to aid in the detection of diabetes retinopathy, AMD, retinal detachment, glaucoma, and ROP, among other ocular disorders. Predictive models for retinal disease prognosis and treatment are also being validated. SUMMARY Deep learning systems have begun to demonstrate a reliable level of diagnostic accuracy equal or better to human graders for narrow image recognition tasks. However, challenges regarding the use of deep learning systems in ophthalmology remain. These include trust of unsupervised learning systems and the limited ability to recognize broad ranges of disorders.
-
4.
Review of Various Tasks Performed in the Preprocessing Phase of a Diabetic Retinopathy Diagnosis System.
Ashraf, MN, Hussain, M, Habib, Z
Current medical imaging. 2020;(4):397-426
Abstract
Diabetic Retinopathy (DR) is a major cause of blindness in diabetic patients. The increasing population of diabetic patients and difficulty to diagnose it at an early stage are limiting the screening capabilities of manual diagnosis by ophthalmologists. Color fundus images are widely used to detect DR lesions due to their comfortable, cost-effective and non-invasive acquisition procedure. Computer Aided Diagnosis (CAD) of DR based on these images can assist ophthalmologists and help in saving many sight years of diabetic patients. In a CAD system, preprocessing is a crucial phase, which significantly affects its performance. Commonly used preprocessing operations are the enhancement of poor contrast, balancing the illumination imbalance due to the spherical shape of a retina, noise reduction, image resizing to support multi-resolution, color normalization, extraction of a field of view (FOV), etc. Also, the presence of blood vessels and optic discs makes the lesion detection more challenging because these two artifacts exhibit specific attributes, which are similar to those of DR lesions. Preprocessing operations can be broadly divided into three categories: 1) fixing the native defects, 2) segmentation of blood vessels, and 3) localization and segmentation of optic discs. This paper presents a review of the state-of-the-art preprocessing techniques related to three categories of operations, highlighting their significant aspects and limitations. The survey is concluded with the most effective preprocessing methods, which have been shown to improve the accuracy and efficiency of the CAD systems.
-
5.
Ophthalmic diagnosis using deep learning with fundus images - A critical review.
Sengupta, S, Singh, A, Leopold, HA, Gulati, T, Lakshminarayanan, V
Artificial intelligence in medicine. 2020;:101758
Abstract
An overview of the applications of deep learning for ophthalmic diagnosis using retinal fundus images is presented. We describe various retinal image datasets that can be used for deep learning purposes. Applications of deep learning for segmentation of optic disk, optic cup, blood vessels as well as detection of lesions are reviewed. Recent deep learning models for classification of diseases such as age-related macular degeneration, glaucoma, and diabetic retinopathy are also discussed. Important critical insights and future research directions are given.
-
6.
3D Retinal Vessel Density Mapping With OCT-Angiography.
Sarabi, MS, Khansari, MM, Zhang, J, Kushner-Lenhoff, S, Gahm, JK, Qiao, Y, Kashani, AH, Shi, Y
IEEE journal of biomedical and health informatics. 2020;(12):3466-3479
-
-
Free full text
-
Abstract
Optical Coherence Tomography Angiography (OCTA) is a novel, non-invasive imaging modality of retinal capillaries at micron resolution. Recent studies have correlated macular OCTA vascular measures with retinal disease severity and supported their use as a diagnostic tool. However, these measurements mostly rely on a few summary statistics in retinal layers or regions of interest in the two-dimensional (2D) en face projection images. To enable 3D and localized comparisons of retinal vasculature between longitudinal scans and across populations, we develop a novel approach for mapping retinal vessel density from OCTA images. We first obtain a high-quality 3D representation of OCTA-based vessel networks via curvelet-based denoising and optimally oriented flux (OOF). Then, an effective 3D retinal vessel density mapping method is proposed. In this framework, a vessel density image (VDI) is constructed by diffusing the vessel mask derived from OOF-based analysis to the entire image volume. Subsequently, we utilize a non-linear, 3D OCT image registration method to provide localized comparisons of retinal vasculature across subjects. In our experimental results, we demonstrate an application of our method for longitudinal qualitative analysis of two pathological subjects with edema during the course of clinical care. Additionally, we quantitatively validate our method on synthetic data with simulated capillary dropout, a dataset obtained from a normal control (NC) population divided into two age groups and a dataset obtained from patients with diabetic retinopathy (DR). Our results show that we can successfully detect localized vascular changes caused by simulated capillary loss, normal aging, and DR pathology even in presence of edema. These results demonstrate the potential of the proposed framework in localized detection of microvascular changes and monitoring retinal disease progression.
-
7.
UD-MIL: Uncertainty-Driven Deep Multiple Instance Learning for OCT Image Classification.
Wang, X, Tang, F, Chen, H, Luo, L, Tang, Z, Ran, AR, Cheung, CY, Heng, PA
IEEE journal of biomedical and health informatics. 2020;(12):3431-3442
Abstract
Deep learning has achieved remarkable success in the optical coherence tomography (OCT) image classification task with substantial labelled B-scan images available. However, obtaining such fine-grained expert annotations is usually quite difficult and expensive. How to leverage the volume-level labels to develop a robust classifier is very appealing. In this paper, we propose a weakly supervised deep learning framework with uncertainty estimation to address the macula-related disease classification problem from OCT images with the only volume-level label being available. First, a convolutional neural network (CNN) based instance-level classifier is iteratively refined by using the proposed uncertainty-driven deep multiple instance learning scheme. To our best knowledge, we are the first to incorporate the uncertainty evaluation mechanism into multiple instance learning (MIL) for training a robust instance classifier. The classifier is able to detect suspicious abnormal instances and abstract the corresponding deep embedding with high representation capability simultaneously. Second, a recurrent neural network (RNN) takes instance features from the same bag as input and generates the final bag-level prediction by considering the individually local instance information and globally aggregated bag-level representation. For more comprehensive validation, we built two large diabetic macular edema (DME) OCT datasets from different devices and imaging protocols to evaluate the efficacy of our method, which are composed of 30,151 B-scans in 1,396 volumes from 274 patients (Heidelberg-DME dataset) and 38,976 B-scans in 3,248 volumes from 490 patients (Triton-DME dataset), respectively. We compare the proposed method with the state-of-the-art approaches, and experimentally demonstrate that our method is superior to alternative methods, achieving volume-level accuracy, F1-score and area under the receiver operating characteristic curve (AUC) of 95.1%, 0.939 and 0.990 on Heidelberg-DME and those of 95.1%, 0.935 and 0.986 on Triton-DME, respectively. Furthermore, the proposed method also yields competitive results on another public age-related macular degeneration OCT dataset, indicating the high potential as an effective screening tool in the clinical practice.
-
8.
Quantitative CEST imaging of amide proton transfer in acute ischaemic stroke.
Msayib, Y, Harston, GWJ, Tee, YK, Sheerin, F, Blockley, NP, Okell, TW, Jezzard, P, Kennedy, J, Chappell, MA
NeuroImage. Clinical. 2019;:101833
Abstract
BACKGROUND Amide proton transfer (APT) imaging may help identify the ischaemic penumbra in stroke patients, the classical definition of which is a region of tissue around the ischaemic core that is hypoperfused and metabolically stressed. Given the potential of APT imaging to complement existing imaging techniques to provide clinically-relevant information, there is a need to develop analysis techniques that deliver a robust and repeatable APT metric. The challenge to accurate quantification of an APT metric has been the heterogeneous in-vivo environment of human tissue, which exhibits several confounding magnetisation transfer effects including spectrally-asymmetric nuclear Overhauser effects (NOEs). The recent literature has introduced various model-free and model-based approaches to analysis that seek to overcome these limitations. OBJECTIVES The objective of this work was to compare quantification techniques for CEST imaging that specifically separate APT and NOE effects for application in the clinical setting. Towards this end a methodological comparison of different CEST quantification techniques was undertaken in healthy subjects, and around clinical endpoints in a cohort of acute stroke patients. METHODS MRI data from 12 patients presenting with ischaemic stroke were retrospectively analysed. Six APT quantification techniques, comprising model-based and model-free techniques, were compared for repeatability and ability for APT to distinguish pathological tissue in acute stroke. RESULTS Robustness analysis of six quantification techniques indicated that the multi-pool model-based technique had the smallest contrast between grey and white matter (2%), whereas model-free techniques exhibited the highest contrast (>30%). Model-based techniques also exhibited the lowest spatial variability, of which 4-pool APTR∗ was by far the most uniform (10% coefficient of variation, CoV), followed by 3-pool analysis (20%). Four-pool analysis yielded the highest ischaemic core contrast-to-noise ratio (0.74). Four-pool modelling of APT effects was more repeatable (3.2% CoV) than 3-pool modelling (4.6% CoV), but this appears to come at the cost of reduced contrast between infarct growth tissue and normal tissue. CONCLUSION The multi-pool measures performed best across the analyses of repeatability, spatial variability, contrast-to-noise ratio, and grey matter-white matter contrast, and might therefore be more suitable for use in clinical imaging of acute stroke. Addition of a fourth pool that separates NOEs and semisolid effects appeared to be more biophysically accurate and provided better separation of the APT signal compared to the 3-pool equivalent, but this improvement appeared be accompanied by reduced contrast between infarct growth tissue and normal tissue.
-
9.
Deep Learning Predicts OCT Measures of Diabetic Macular Thickening From Color Fundus Photographs.
Arcadu, F, Benmansour, F, Maunz, A, Michon, J, Haskova, Z, McClintock, D, Adamis, AP, Willis, JR, Prunotto, M
Investigative ophthalmology & visual science. 2019;(4):852-857
Abstract
PURPOSE To develop deep learning (DL) models for the automatic detection of optical coherence tomography (OCT) measures of diabetic macular thickening (MT) from color fundus photographs (CFPs). METHODS Retrospective analysis on 17,997 CFPs and their associated OCT measurements from the phase 3 RIDE/RISE diabetic macular edema (DME) studies. DL with transfer-learning cascade was applied on CFPs to predict time-domain OCT (TD-OCT)-equivalent measures of MT, including central subfield thickness (CST) and central foveal thickness (CFT). MT was defined by using two OCT cutoff points: 250 μm and 400 μm. A DL regression model was developed to directly quantify the actual CFT and CST from CFPs. RESULTS The best DL model was able to predict CST ≥ 250 μm and CFT ≥ 250 μm with an area under the curve (AUC) of 0.97 (95% confidence interval [CI], 0.89-1.00) and 0.91 (95% CI, 0.76-0.99), respectively. To predict CST ≥ 400 μm and CFT ≥ 400 μm, the best DL model had an AUC of 0.94 (95% CI, 0.82-1.00) and 0.96 (95% CI, 0.88-1.00), respectively. The best deep convolutional neural network regression model to quantify CST and CFT had an R2 of 0.74 (95% CI, 0.49-0.91) and 0.54 (95% CI, 0.20-0.87), respectively. The performance of the DL models declined when the CFPs were of poor quality or contained laser scars. CONCLUSIONS DL is capable of predicting key quantitative TD-OCT measurements related to MT from CFPs. The DL models presented here could enhance the efficiency of DME diagnosis in tele-ophthalmology programs, promoting better visual outcomes. Future research is needed to validate DL algorithms for MT in the real-world.
-
10.
Diabetic retinopathy detection through novel tetragonal local octa patterns and extreme learning machines.
Nazir, T, Irtaza, A, Shabbir, Z, Javed, A, Akram, U, Mahmood, MT
Artificial intelligence in medicine. 2019;:101695
Abstract
Diabetic retinopathy (DR) is an eye disease that victimize the people suffering from diabetes from many years. The severe form of DR results in form of the blindness that can initially be controlled by the DR-screening oriented treatment. The effective screening programs require the trained human resource that manually grade the fundus images to understand the severity of the disease. But due to the complexity of this process, and the insufficient number of the trained workers, the precise manual grading is an expensive process. The CAD-based solutions try to address these limitations but most of the existing DR detection systems are as evaluated over small sets and become ineffective when applied in real scenarios. Therefore, in this paper we proposed a novel technique to precisely detect the various stages of the DR by extending the research of the content-based image retrieval domain. To achieve the human-level performance over the large-scale DR-datasets (i.e. Kaggle-DR), the fundus images are represented by the novel tetragonal local octa pattern (T-LOP) features, that are then classified through the extreme learning machine (ELM). To justify the significance of the method, the proposed scheme is compared against several state-of-the-art methods including the deep learning-based methods over four DR-datasets of variational lengths (i.e. Kaggle-DR, DRIVE, Review-DB, STARE). The experimental results confirm the significance of the DR-detection scheme to serve as a stand-alone solution for providing the precise information of the severity of the DR in an efficient manner.