-
Journal of Clinical Medicine May 2024: To design a novel anomaly detection and localization approach using artificial intelligence methods using optical coherence tomography (OCT) scans for retinal...
: To design a novel anomaly detection and localization approach using artificial intelligence methods using optical coherence tomography (OCT) scans for retinal diseases. : High-resolution OCT scans from the publicly available Kaggle dataset and a local dataset were used by four state-of-the-art self-supervised frameworks. The backbone model of all the frameworks was a pre-trained convolutional neural network (CNN), which enabled the extraction of meaningful features from OCT images. Anomalous images included choroidal neovascularization (CNV), diabetic macular edema (DME), and the presence of drusen. Anomaly detectors were evaluated by commonly accepted performance metrics, including area under the receiver operating characteristic curve, F1 score, and accuracy. : A total of 25,315 high-resolution retinal OCT slabs were used for training. Test and validation sets consisted of 968 and 4000 slabs, respectively. The best performing across all anomaly detectors had an area under the receiver operating characteristic of 0.99. All frameworks were shown to achieve high performance and generalize well for the different retinal diseases. Heat maps were generated to visualize the quality of the frameworks' ability to localize anomalous areas of the image. : This study shows that with the use of pre-trained feature extractors, the frameworks tested can generalize to the domain of retinal OCT scans and achieve high image-level ROC-AUC scores. The localization results of these frameworks are promising and successfully capture areas that indicate the presence of retinal pathology. Moreover, such frameworks have the potential to uncover new biomarkers that are difficult for the human eye to detect. Frameworks for anomaly detection and localization can potentially be integrated into clinical decision support and automatic screening systems that will aid ophthalmologists in patient diagnosis, follow-up, and treatment design. This work establishes a solid basis for further development of automated anomaly detection frameworks for clinical use.
PubMed: 38892804
DOI: 10.3390/jcm13113093 -
Investigative Ophthalmology & Visual... Jun 2024In age-related macular degeneration (AMD), choriocapillaris flow deficits (CCFDs) under soft drusen can be measured using established compensation strategies. This study...
PURPOSE
In age-related macular degeneration (AMD), choriocapillaris flow deficits (CCFDs) under soft drusen can be measured using established compensation strategies. This study investigated whether CCFDs can be quantified under calcified drusen (CaD).
METHODS
CCFDs were measured in normal eyes (n = 30) and AMD eyes with soft drusen (n = 30) or CaD (n = 30). CCFD density masks were generated to highlight regions with higher CCFDs. Masks were also generated for soft drusen and CaD based on both structural en face OCT images and corresponding B-scans. Dice similarity coefficients were calculated between the CCFD density masks and both the soft drusen and CaD masks. A phantom experiment was conducted to simulate the impact of light scattering that arises from CaD.
RESULTS
Area measurements of CCFDs were highly correlated with those of CaD but not soft drusen, suggesting an association between CaD and underlying CCFDs. However, unlike soft drusen, the detected optical coherence tomography (OCT) signals underlying CaD did not arise from the defined CC layer but were artifacts caused by the multiple scattering property of CaD. Phantom experiments showed that the presence of highly scattering material similar to the contents of CaD caused an artifactual scattering tail that falsely generated a signal in the CC structural layer but the underlying flow could not be detected. Similarly, CaD also caused an artifactual scattering tail and prevented the penetration of light into the choroid, resulting in en face hypotransmission defects and an inability to detect blood flow within the choriocapillaris. Upon resolution of the CaD, the CC perfusion became detectable.
CONCLUSIONS
The high scattering property of CaD leads to a scattering tail under these drusen that gives the illusion of a quantifiable optical coherence tomography angiography signal, but this signal does not contain the angiographic information required to assess CCFDs. For this reason, CCFDs cannot be reliably measured under CaD, and CaD must be identified and excluded from macular CCFD measurements.
Topics: Humans; Tomography, Optical Coherence; Choroid; Retinal Drusen; Female; Aged; Male; Fluorescein Angiography; Regional Blood Flow; Calcinosis; Aged, 80 and over; Macular Degeneration; Middle Aged; Phantoms, Imaging; Fundus Oculi
PubMed: 38884553
DOI: 10.1167/iovs.65.6.26 -
Ophthalmology. Retina Jun 2024To investigate the relationships between contrast sensitivity (CS), choriocapillaris perfusion and other structural optical coherence tomography (OCT) biomarkers in dry...
PURPOSE
To investigate the relationships between contrast sensitivity (CS), choriocapillaris perfusion and other structural optical coherence tomography (OCT) biomarkers in dry age-related macular degeneration (AMD).
DESIGN
Cross-sectional, observational study.
PARTICIPANTS
One hundred AMD eyes (22 early, 52 intermediate and 26 late) from 74 patients and 45 control eyes from 37 age-similar subjects.
METHODS
All participants had visual acuity (VA) assessment, quantitative contrast sensitivity function (qCSF) testing, macular OCT, and 6x6-mm swept-source OCT angiography (OCTA) scans on the same day. OCT volumes were analyzed for subretinal drusenoid deposits and hyporeflective drusen cores, and to measure thickness of the outer nuclear layer (ONL). OCTA scans were utilized to calculate drusen volume, inner choroid flow deficit percentage (IC-FD%), and to measure the area of choroidal hypertransmission defects (HTD). IC-FD% was measured from a 16 μm-thick choriocapillaris slab after compensation and binarization with Phansalkar's method. Generalized linear mixed-effects models were used to evaluate the associations between functional and structural variables.
MAIN OUTCOME MEASURES
To explore the associations between qCSF-measured CS, ICFD% and various AMD imaging biomarkers.
RESULTS
AMD exhibited significantly reduced qCSF metrics eyes across all stages compared to controls. Univariate analysis revealed significant associations between various imaging biomarkers, reduced qCSF metrics and VA in both groups. Multivariate analysis confirmed that higher IC-FD% in the central 5 mm was significantly associated with decreases in all qCSF metrics in AMD eyes (β= -0.74 to -0.25, all p<0.05), but not with VA (p>0.05). ONL thickness in the central 3 mm correlated with both VA (β= 2.85, p<0.001) and several qCSF metrics (β= 0.01-0.90, all p<0.05), especially in AMD eyes. Further, larger HTD areas were associated with decreased VA (β=-0.89, p<0.001) and reduced CS at low-intermediate frequencies across AMD stages (β= -0.30 to -0.29, p<0.001).
CONCLUSIONS
The significant association between IC-FD% in the central 5 mm and qCSF-measured CS reinforces the hypothesis that decreased macular choriocapillaris perfusion contributes to visual function changes in AMD, which are more pronounced in CS than in VA.
PubMed: 38878897
DOI: 10.1016/j.oret.2024.06.005 -
Computer Methods and Programs in... May 2024Optical coherence tomography (OCT) has ushered in a transformative era in the domain of ophthalmology, offering non-invasive imaging with high resolution for ocular... (Review)
Review
BACKGROUND AND OBJECTIVES
Optical coherence tomography (OCT) has ushered in a transformative era in the domain of ophthalmology, offering non-invasive imaging with high resolution for ocular disease detection. OCT, which is frequently used in diagnosing fundamental ocular pathologies, such as glaucoma and age-related macular degeneration (AMD), plays an important role in the widespread adoption of this technology. Apart from glaucoma and AMD, we will also investigate pertinent pathologies, such as epiretinal membrane (ERM), macular hole (MH), macular dystrophy (MD), vitreomacular traction (VMT), diabetic maculopathy (DMP), cystoid macular edema (CME), central serous chorioretinopathy (CSC), diabetic macular edema (DME), diabetic retinopathy (DR), drusen, glaucomatous optic neuropathy (GON), neovascular AMD (nAMD), myopia macular degeneration (MMD) and choroidal neovascularization (CNV) diseases. This comprehensive review examines the role that OCT-derived images play in detecting, characterizing, and monitoring eye diseases.
METHOD
The 2020 PRISMA guideline was used to structure a systematic review of research on various eye conditions using machine learning (ML) or deep learning (DL) techniques. A thorough search across IEEE, PubMed, Web of Science, and Scopus databases yielded 1787 publications, of which 1136 remained after removing duplicates. Subsequent exclusion of conference papers, review papers, and non-open-access articles reduced the selection to 511 articles. Further scrutiny led to the exclusion of 435 more articles due to lower-quality indexing or irrelevance, resulting in 76 journal articles for the review.
RESULTS
During our investigation, we found that a major challenge for ML-based decision support is the abundance of features and the determination of their significance. In contrast, DL-based decision support is characterized by a plug-and-play nature rather than relying on a trial-and-error approach. Furthermore, we observed that pre-trained networks are practical and especially useful when working on complex images such as OCT. Consequently, pre-trained deep networks were frequently utilized for classification tasks. Currently, medical decision support aims to reduce the workload of ophthalmologists and retina specialists during routine tasks. In the future, it might be possible to create continuous learning systems that can predict ocular pathologies by identifying subtle changes in OCT images.
PubMed: 38861878
DOI: 10.1016/j.cmpb.2024.108253 -
PloS One 2024Age-related macular degeneration (AMD) is an eye disease that leads to the deterioration of the central vision area of the eye and can gradually result in vision loss in...
Age-related macular degeneration (AMD) is an eye disease that leads to the deterioration of the central vision area of the eye and can gradually result in vision loss in elderly individuals. Early identification of this disease can significantly impact patient treatment outcomes. Furthermore, given the increasing elderly population globally, the importance of automated methods for rapidly monitoring at-risk individuals and accurately diagnosing AMD is growing daily. One standard method for diagnosing AMD is using optical coherence tomography (OCT) images as a non-invasive imaging technology. In recent years, numerous deep neural networks have been proposed for the classification of OCT images. Utilizing pre-trained neural networks can speed up model deployment in related tasks without compromising accuracy. However, most previous methods overlook the feasibility of leveraging pre-existing trained networks to search for an optimal architecture for AMD staging on a new target dataset. In this study, our objective was to achieve an optimal architecture in the efficiency-accuracy trade-off for classifying retinal OCT images. To this end, we employed pre-trained medical vision transformer (MedViT) models. MedViT combines convolutional and transformer neural networks, explicitly designed for medical image classification. Our approach involved pre-training two distinct MedViT models on a source dataset with labels identical to those in the target dataset. This pre-training was conducted in a supervised manner. Subsequently, we evaluated the performance of the pre-trained MedViT models for classifying retinal OCT images from the target Noor Eye Hospital (NEH) dataset into the normal, drusen, and choroidal neovascularization (CNV) classes in zero-shot settings and through five-fold cross-validation. Then, we proposed a stitching approach to search for an optimal model from two MedViT family models. The proposed stitching method is an efficient architecture search algorithm known as stitchable neural networks. Stitchable neural networks create a candidate model in search space for each pair of stitchable layers by inserting a linear layer between them. A pair of stitchable layers consists of layers, each selected from one input model. While stitchable neural networks had previously been tested on more extensive and general datasets, this study demonstrated that stitching networks could also be helpful in smaller medical datasets. The results of this approach indicate that when pre-trained models were available for OCT images from another dataset, it was possible to achieve a model in 100 epochs with an accuracy of over 94.9% in classifying images from the NEH dataset. The results of this study demonstrate the efficacy of stitchable neural networks as a fine-tuning method for OCT image classification. This approach not only leads to higher accuracy but also considers architecture optimization at a reasonable computational cost.
Topics: Tomography, Optical Coherence; Humans; Macular Degeneration; Retina; Neural Networks, Computer; Aged; Algorithms
PubMed: 38837967
DOI: 10.1371/journal.pone.0304943 -
Ophthalmology Science 2024To elucidate the clinical characteristics and progression rates of pachychoroid and conventional geographic atrophy (GA).
PURPOSE
To elucidate the clinical characteristics and progression rates of pachychoroid and conventional geographic atrophy (GA).
DESIGN
Retrospective, multicenter, observational study.
PARTICIPANTS
A total of 173 eyes from 173 patients (38 eyes with pachychoroid GA and 135 with conventional GA) from 6 university hospitals in Japan were included. All patients were Japanese, aged ≥50 years and with fundus autofluorescence images having analyzable image quality. A total of 101 eyes (22 with pachychoroid GA and 79 with conventional GA) were included in the follow-up group.
METHODS
The studied eyes were classified as having pachychoroid or conventional GA; the former was diagnosed if the eye had features of pachychoroid and no drusen. The GA area was semiautomatically measured on fundus autofluorescence images, and the GA progression rate was calculated for the follow-up group. Multivariable linear regression analysis was used to determine whether the rate of GA progression was associated with GA subtype.
MAIN OUTCOME MEASURES
Clinical characteristics and progression rates of pachychoroid and conventional GA.
RESULTS
The pachychoroid GA group was significantly younger (70.3 vs. 78.7 years; < 0.001), more male-dominant (89.5 vs. 55.6%; < 0.001), and had better best-corrected visual acuity (0.15 vs. 0.40 in logarithm of the minimum angle of resolution; = 0.002), thicker choroid (312.4 vs. 161.6 μm; < 0.001), higher rate of unifocal GA type (94.7 vs. 49.6%; < 0.001), and smaller GA area (0.59 vs. 3.76 mm < 0.001) than the conventional GA group. In the follow-up group, the mean GA progression rate (square-root transformation) was significantly lower in the pachychoroid GA group than in the conventional GA group (0.11 vs. 0.27 mm/year; < 0.001).
CONCLUSIONS
Demographic and ocular characteristics differed between GA subtypes. The progression rate of pachychoroid GA, adjusted for age and baseline GA area, was significantly lower than that of conventional GA. Japanese patients with conventional GA showed characteristics and progression rates similar to those in White populations. Some characteristics of GA in Japanese population differ from those in Waucasian populations, which may be due to the inclusion of pachychoroid GA.
FINANCIAL DISCLOSURES
Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
PubMed: 38827489
DOI: 10.1016/j.xops.2024.100528 -
Cell Death & Disease Jun 2024Drusen, the yellow deposits under the retina, are composed of lipids and proteins, and represent a hallmark of age-related macular degeneration (AMD). Lipid droplets are...
Drusen, the yellow deposits under the retina, are composed of lipids and proteins, and represent a hallmark of age-related macular degeneration (AMD). Lipid droplets are also reported in the retinal pigment epithelium (RPE) from AMD donor eyes. However, the mechanisms underlying these disease phenotypes remain elusive. Previously, we showed that Pgc-1α repression, combined with a high-fat diet (HFD), induce drastic AMD-like phenotypes in mice. We also reported increased PGC-1α acetylation and subsequent deactivation in the RPE derived from AMD donor eyes. Here, through a series of in vivo and in vitro experiments, we sought to investigate the molecular mechanisms by which PGC-1α repression could influence RPE and retinal function. We show that PGC-1α plays an important role in RPE and retinal lipid metabolism and function. In mice, repression of Pgc-1α alone induced RPE and retinal degeneration and drusen-like deposits. In vitro inhibition of PGC1A by CRISPR-Cas9 gene editing in human RPE (ARPE19- PGC1A KO) affected the expression of genes responsible for lipid metabolism, fatty acid β-oxidation (FAO), fatty acid transport, low-density lipoprotein (LDL) uptake, cholesterol esterification, cholesterol biosynthesis, and cholesterol efflux. Moreover, inhibition of PGC1A in RPE cells caused lipid droplet accumulation and lipid peroxidation. ARPE19-PGC1A KO cells also showed reduced mitochondrial biosynthesis, impaired mitochondrial dynamics and activity, reduced antioxidant enzymes, decreased mitochondrial membrane potential, loss of cardiolipin, and increased susceptibility to oxidative stress. Our data demonstrate the crucial role of PGC-1α in regulating lipid metabolism. They provide new insights into the mechanisms involved in lipid and drusen accumulation in the RPE and retina during aging and AMD, which may pave the way for developing novel therapeutic strategies targeting PGC-1α.
Topics: Retinal Pigment Epithelium; Peroxisome Proliferator-Activated Receptor Gamma Coactivator 1-alpha; Animals; Lipid Metabolism; Humans; Mice; Lipid Droplets; Macular Degeneration; Mice, Inbred C57BL; Mitochondria; Male; Oxidative Stress
PubMed: 38824126
DOI: 10.1038/s41419-024-06762-y -
Clinical & Experimental Ophthalmology May 2024To examine whether the clinical performance of predicting late age-related macular degeneration (AMD) development is improved through using multimodal imaging (MMI)...
BACKGROUND
To examine whether the clinical performance of predicting late age-related macular degeneration (AMD) development is improved through using multimodal imaging (MMI) compared to using colour fundus photography (CFP) alone, and how this compares with a basic prediction model using well-established AMD risk factors.
METHODS
Individuals with AMD in this study underwent MMI, including optical coherence tomography (OCT), fundus autofluorescence, near-infrared reflectance and CFP at baseline, and then at 6-monthly intervals for 3-years to determine MMI-defined late AMD development. Four retinal specialists independently assessed the likelihood that each eye at baseline would progress to MMI-defined late AMD over 3-years with CFP, and then with MMI. Predictive performance with CFP and MMI were compared to each other, and to a basic prediction model using age, presence of pigmentary abnormalities, and OCT-based drusen volume.
RESULTS
The predictive performance of the clinicians using CFP [AUC = 0.75; 95% confidence interval (CI) = 0.68-0.82] improved when using MMI (AUC = 0.79; 95% CI = 0.72-0.85; p = 0.034). However, a basic prediction model outperformed clinicians using either CFP or MMI (AUC = 0.85; 95% CI = 0.78-91; p ≤ 0.002).
CONCLUSIONS
Clinical performance for predicting late AMD development was improved by using MMI compared to CFP. However, a basic prediction model using well-established AMD risk factors outperformed retinal specialists, suggesting that such a model could further improve personalised counselling and monitoring of individuals with the early stages of AMD in clinical practice.
PubMed: 38812454
DOI: 10.1111/ceo.14405 -
Lasers in Medical Science May 2024Classifying retinal diseases is a complex problem because the early problematic areas of retinal disorders are quite small and conservative. In recent years, Transformer... (Comparative Study)
Comparative Study
Classifying retinal diseases is a complex problem because the early problematic areas of retinal disorders are quite small and conservative. In recent years, Transformer architectures have been successfully applied to solve various retinal related health problems. Age-related macular degeneration (AMD) and diabetic macular edema (DME), two prevalent retinal diseases, can cause partial or total blindness. Diseases therefore require an early and accurate detection. In this study, we proposed Vision Transformer (ViT), Tokens-To-Token Vision Transformer (T2T-ViT) and Mobile Vision Transformer (Mobile-ViT) algorithms to detect choroidal neovascularization (CNV), drusen, and diabetic macular edema (DME), and normal using optical coherence tomography (OCT) images. The predictive accuracies of ViT, T2T-ViT and Mobile-ViT achieved on the dataset for the classification of OCT images are 95.14%, 96.07% and 99.17% respectively. Experimental results obtained from ViT approaches showed that Mobile-ViT have superior performance with regard to classification accuracy in comparison with the others. Overall, it has been observed that ViT architectures have the capacity to classify with high accuracy in the diagnosis of retinal diseases.
Topics: Tomography, Optical Coherence; Humans; Diabetic Retinopathy; Choroidal Neovascularization; Macular Edema; Algorithms; Retinal Drusen; Retina
PubMed: 38797751
DOI: 10.1007/s10103-024-04089-w -
Investigative Ophthalmology & Visual... May 2024To explore the association between the genetics of age-related macular degeneration (AMD) and extramacular drusen (EMD) in patients with and without AMD.
PURPOSE
To explore the association between the genetics of age-related macular degeneration (AMD) and extramacular drusen (EMD) in patients with and without AMD.
METHODS
We included 1753 eyes (912 subjects) with phenotypic characterization regarding AMD and EMD. Genetic sequencing and the genetic risk score (GRS) for AMD were performed according to the EYE-RISK consortium methodology. To test for differences in the GRS from EMD cases, AMD cases, and controls, a clustered Wilcoxon rank-sum test was used. The association of AMD, EMD, and the GRS was evaluated using logistic regression models adjusted for age and sex. Individual associations of common risk variants for AMD with EMD were explored.
RESULTS
EMD were found in 755 eyes: 252 (14.4%) with AMD and 503 (28.7%) without. In total, 122 eyes (7.0%) had only AMD, and 876 (50.0%) were controls. EMD were strongly associated with AMD (odds ratio [OR], 3.333; 95% confidence interval [CI], 2.356-4.623; P < 0.001). The GRS was associated with an increased risk of AMD (OR, 1.416; 95% CI, 1.218-1.646; P < 0.001) but not with EMD. Individually, the common risk variants ARMS2 rs10490924 (P = 0.042), C3 rs2230199 (P = 0.042), and CETP rs5817082 (P = 0.042) were associated with EMD, after adjustment for AMD, sex, and age.
CONCLUSIONS
We found a strong association between EMD and AMD, suggesting a common pathogenesis. The GRS for AMD was not associated with EMD, but a partially overlapping genetic basis was suggested when assessing individual risk variants. We propose that EMD per se do not represent an increase in the global genetic risk for AMD.
Topics: Humans; Female; Male; Macular Degeneration; Retinal Drusen; Aged; Middle Aged; Aged, 80 and over; Genetic Predisposition to Disease; Risk Factors; Polymorphism, Single Nucleotide; Proteins
PubMed: 38776116
DOI: 10.1167/iovs.65.5.35