Supervised Machine Learning Based Multi-Task Artificial Intelligence Classification of Retinopathies

A Rapid Review of Interventions to Improve Care for People Who Are Medically Underserved with Multiple Sclerosis, Diabetic Retinopathy, and Lung Cancer

In the United States, patients with chronic conditions experience disparities in health outcomes across the care continuum. Among patients with multiple sclerosis, diabetic retinopathy, and lung cancer, there is a lack of evidence summarizing interventions to improve care and decrease these disparities. The aim of this rapid literature review was to identify interventions among patients with these chronic conditions to improve health and reduce disparities in screening, diagnosis, access to treatment and specialists, adherence, and retention in care. Using structured search terms in PubMed and Web of Science, we completed a rapid review of studies published in the prior five years conducted in the United States on our subject of focus. We screened the retrieved articles for inclusion and extracted data using a standard spreadsheet. The data were synthesized across clinical conditions and summarized. Screening was the most common point in the care continuum with documented interventions. Most studies we identified addressed interventions for patients with lung cancer, with half as many studies identified for patients with diabetic retinopathy, and few studies identified for patients with multiple sclerosis. Almost two-thirds of the studies focused on patients who identify as Black, Indigenous, or people of color. Interventions with evidence evaluating implementation in multiple conditions included telemedicine, mobile clinics, and insurance subsidies, or expansion. Despite documented disparities and a focus on health equity, a paucity of evidence exists on interventions that improve health outcomes among patients who are medically underserved with multiple sclerosis, diabetic retinopathy, and lung cancer.

DISCOVER: 2-D multiview summarization of Optical Coherence Tomography Angiography for automatic diabetic retinopathy diagnosis

Diabetic Retinopathy (DR), an ocular complication of diabetes, is a leading cause of blindness worldwide. Traditionally, DR is monitored using Color Fundus Photography (CFP), a widespread 2-D imaging modality. However, DR classifications based on CFP have poor predictive power, resulting in suboptimal DR management. Optical Coherence Tomography Angiography (OCTA) is a recent 3-D imaging modality offering enhanced structural and functional information (blood flow) with a wider field of view. This paper investigates automatic DR severity assessment using 3-D OCTA. A straightforward solution to this task is a 3-D neural network classifier. However, 3-D architectures have numerous parameters and typically require many training samples. A lighter solution consists in using 2-D neural network classifiers processing 2-D en-face (or frontal) projections and/or 2-D cross-sectional slices. Such an approach mimics the way ophthalmologists analyze OCTA acquisitions: (1) en-face flow maps are often used to detect avascular zones and neovascularization, and (2) cross-sectional slices are commonly analyzed to detect macular edemas, for instance. However, arbitrary data reduction or selection might result in information loss. Two complementary strategies are thus proposed to optimally summarize OCTA volumes with 2-D images: (1) a parametric en-face projection optimized through deep learning and (2) a cross-sectional slice selection process controlled through gradient-based attribution. The full summarization and DR classification pipeline is trained from end to end. The automatic 2-D summary can be displayed in a viewer or printed in a report to support the decision. We show that the proposed 2-D summarization and classification pipeline outperforms direct 3-D classification with the advantage of improved interpretability.

Optical Coherence Tomography Angiography as a Diagnostic Tool for Diabetic Retinopathy

Diabetic retinopathy (DR) is the most common microvascular complication of diabetes mellitus, leading to visual impairment if left untreated. This review discusses the use of optical coherence tomography angiography (OCTA) as a diagnostic tool for the early detection and management of DR. OCTA is a fast, non-invasive, non-contact test that enables the detailed visualisation of the macular microvasculature in different plexuses. OCTA offers several advantages over fundus fluorescein angiography (FFA), notably offering quantitative data. OCTA is not without limitations, including the requirement for careful interpretation of artefacts and the limited region of interest that can be captured currently. We explore how OCTA has been instrumental in detecting early microvascular changes that precede clinical signs of DR. We also discuss the application of OCTA in the diagnosis and management of various stages of DR, including non-proliferative diabetic retinopathy (NPDR), proliferative diabetic retinopathy (PDR), diabetic macular oedema (DMO), diabetic macular ischaemia (DMI), and pre-diabetes. Finally, we discuss the future role of OCTA and how it may be used to enhance the clinical outcomes of DR.

Ocular microvascular complications in diabetic retinopathy: insights from machine learning

INTRODUCTION

Diabetic retinopathy (DR) is a leading cause of preventable blindness among working-age adults, primarily driven by ocular microvascular complications from chronic hyperglycemia. Comprehending the complex relationship between microvascular changes in the eye and disease progression poses challenges, traditional methods assuming linear or logistical relationships may not adequately capture the intricate interactions between these changes and disease advances. Hence, the aim of this study was to evaluate the microvascular involvement of diabetes mellitus (DM) and non-proliferative DR with the implementation of non-parametric machine learning methods.

RESEARCH DESIGN AND METHODS

We conducted a retrospective cohort study that included optical coherence tomography angiography (OCTA) images collected from a healthy group (196 eyes), a DM no DR group (120 eyes), a mild DR group (71 eyes), and a moderate DR group (66 eyes). We implemented a non-parametric machine learning method for four classification tasks that used parameters extracted from the OCTA images as predictors: DM no DR versus healthy, mild DR versus DM no DR, moderate DR versus mild DR, and any DR versus no DR. SHapley Additive exPlanations values were used to determine the importance of these parameters in the classification.

RESULTS

We found large choriocapillaris flow deficits were the most important for healthy versus DM no DR, and became less important in eyes with mild or moderate DR. The superficial microvasculature was important for the healthy versus DM no DR and mild DR versus moderate DR tasks, but not for the DM no DR versus mild DR task-the stage when deep microvasculature plays an important role. Foveal avascular zone metric was in general less affected, but its involvement increased with worsening DR.

CONCLUSIONS

The findings from this study provide valuable insights into the microvascular involvement of DM and DR, facilitating the development of early detection methods and intervention strategies.

Artificial intelligence and hemodynamic studies in optical coherence tomography angiography for diabetic retinopathy evaluation: A review

Diabetic retinopathy (DR) is a rapidly emerging retinal abnormality worldwide, which can cause significant vision loss by disrupting the vascular structure in the retina. Recently, optical coherence tomography angiography (OCTA) has emerged as an effective imaging tool for diagnosing and monitoring DR. OCTA produces high-quality 3-dimensional images and provides deeper visualization of retinal vessel capillaries and plexuses. The clinical relevance of OCTA in detecting, classifying, and planning therapeutic procedures for DR patients has been highlighted in various studies. Quantitative indicators obtained from OCTA, such as blood vessel segmentation of the retina, foveal avascular zone (FAZ) extraction, retinal blood vessel density, blood velocity, flow rate, capillary vessel pressure, and retinal oxygen extraction, have been identified as crucial hemodynamic features for screening DR using computer-aided systems in artificial intelligence (AI). AI has the potential to assist physicians and ophthalmologists in developing new treatment options. In this review, we explore how OCTA has impacted the future of DR screening and early diagnosis. It also focuses on how analysis methods have evolved over time in clinical trials. The future of OCTA imaging and its continued use in AI-assisted analysis is promising and will undoubtedly enhance the clinical management of DR.

Sensitivity and specificity of optical coherence tomography angiography for diagnosis and classification of diabetic retinopathy; a systematic review and meta-analysis

BACKGROUND

Optical coherence tomography angiography (OCTA) is a noninvasive imaging method that can be used for the staging of diabetic retinopathy. In addition, alterations in OCTA parameters can precede the clinical fundus changes. In this review, we aimed to assess the accuracy of OCTA in diagnosis and staging of diabetic retinopathy.

METHODS

Two independent reviewers participated in the literature search using electronic databases (PubMed, Embase, Cochrane Library Central Register of Controlled Trials, ISI, and Scopus) from inception till December 2020. The heterogeneity of data was assessed by Q statistics, Chi-square test and I2 index.

RESULTS

Forty-four articles published from 2015 to the end of 2020 were included in this meta-analysis. Of these, 27 were case-control studies, 9 were case series, and 8 were cohort studies. In total, 4284 eyes of 3553 patients were assessed in this study. OCTA could differentiate diabetic retinopathy from diabetes without diabetic retinopathy with a sensitivity of 88% (95% CI: 85% to 92%) and specificity of 88% (95% CI: 85% to 91%). In addition, it could differentiate proliferative diabetic retinopathy from non-proliferative diabetic retinopathy with a sensitivity of 91% (95% CI: 86% to 95%) and specificity of 91% (95% CI:86% to 96%). The sensitivity of OCTA for diagnosing diabetic retinopathy was increased by the size of scan (3 × 3 mm: 85%; 6 × 6 mm: 91%, 12 × 12 mm: 96%).

CONCLUSION

OCTA, as a non-invasive method, has acceptable sensitivity and specificity for diagnosis and classification of diabetic retinopathy. A larger scan size is associated with more sensitivity for discriminating diabetic retinopathy.

The effectiveness of interventions for optometric referrals into the hospital eye service: A review

PURPOSE

Ophthalmic services are currently under considerable stress; in the UK, ophthalmology departments have the highest number of outpatient appointments of any department within the National Health Service. Recognising the need for intervention, several approaches have been trialled to tackle the high numbers of false-positive referrals initiated in primary care and seen face to face within the hospital eye service (HES). In this mixed-methods narrative synthesis, we explored interventions based on their clinical impact, cost and acceptability to determine whether they are clinically effective, safe and sustainable. A systematic literature search of PubMed, MEDLINE and CINAHL, guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), was used to identify appropriate studies published between December 2001 and December 2022.

RECENT FINDINGS

A total of 55 studies were reviewed. Four main interventions were assessed, where two studies covered more than one type: training and guidelines (n = 8), referral filtering schemes (n = 32), asynchronous teleophthalmology (n = 13) and synchronous teleophthalmology (n = 5). All four approaches demonstrated effectiveness for reducing false-positive referrals to the HES. There was sufficient evidence for stakeholder acceptance and cost-effectiveness of referral filtering schemes; however, cost comparisons involved assumptions. Referral filtering and asynchronous teleophthalmology reported moderate levels of false-negative cases (2%-20%), defined as discharged patients requiring HES monitoring.

SUMMARY

The effectiveness of interventions varied depending on which outcome and stakeholder was considered. More studies are required to explore stakeholder opinions around all interventions. In order to maximise clinical safety, it may be appropriate to combine more than one approach, such as referral filtering schemes with virtual review of discharged patients to assess the rate of false-negative cases. The implementation of a successful intervention is more complex than a 'one-size-fits-all' approach and there is potential space for newer types of interventions, such as artificial intelligence clinical support systems within the referral pathway.

Artificial intelligence in retinal image analysis: Development, advances, and challenges

Modern advances in diagnostic technologies offer the potential for unprecedented insight into ophthalmic conditions relating to the retina. We discuss the current landscape of artificial intelligence in retina with respect to screening, diagnosis, and monitoring of retinal pathologies such as diabetic retinopathy, diabetic macular edema, central serous chorioretinopathy, and age-related macular degeneration. We review the methods used in these models and evaluate their performance in both research and clinical contexts and discuss potential future directions for investigation, use of multiple imaging modalities in artificial intelligence algorithms, and challenges in the application of artificial intelligence in retinal pathologies.

Comparison and development of machine learning for thalidomide-induced peripheral neuropathy prediction of refractory Crohn’s disease in Chinese population

BACKGROUND

Thalidomide is an effective treatment for refractory Crohn’s disease (CD). However, thalidomide-induced peripheral neuropathy (TiPN), which has a large individual variation, is a major cause of treatment failure. TiPN is rarely predictable and recognized, especially in CD. It is necessary to develop a risk model to predict TiPN occurrence.

AIM

To develop and compare a predictive model of TiPN using machine learning based on comprehensive clinical and genetic variables.

METHODS

A retrospective cohort of 164 CD patients from January 2016 to June 2022 was used to establish the model. The National Cancer Institute Common Toxicity Criteria Sensory Scale (version 4.0) was used to assess TiPN. With 18 clinical features and 150 genetic variables, five predictive models were established and evaluated by the confusion matrix receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), specificity, sensitivity (recall rate), precision, accuracy, and F1 score.

RESULTS

The top-ranking five risk variables associated with TiPN were interleukin-12 rs1353248 [P = 0.0004, odds ratio (OR): 8.983, 95% confidence interval (CI): 2.497-30.90], dose (mg/d, P = 0.002), brain-derived neurotrophic factor (BDNF) rs2030324 (P = 0.001, OR: 3.164, 95%CI: 1.561-6.434), BDNF rs6265 (P = 0.001, OR: 3.150, 95%CI: 1.546-6.073) and BDNF rs11030104 (P = 0.001, OR: 3.091, 95%CI: 1.525-5.960). In the training set, gradient boosting decision tree (GBDT), extremely random trees (ET), random forest, logistic regression and extreme gradient boosting (XGBoost) obtained AUROC values > 0.90 and AUPRC > 0.87. Among these models, XGBoost and GBDT obtained the first two highest AUROC (0.90 and 1), AUPRC (0.98 and 1), accuracy (0.96 and 0.98), precision (0.90 and 0.95), F1 score (0.95 and 0.98), specificity (0.94 and 0.97), and sensitivity (1). In the validation set, XGBoost algorithm exhibited the best predictive performance with the highest specificity (0.857), accuracy (0.818), AUPRC (0.86) and AUROC (0.89). ET and GBDT obtained the highest sensitivity (1) and F1 score (0.8). Overall, compared with other state-of-the-art classifiers such as ET, GBDT and RF, XGBoost algorithm not only showed a more stable performance, but also yielded higher ROC-AUC and PRC-AUC scores, demonstrating its high accuracy in prediction of TiPN occurrence.

CONCLUSION

The powerful XGBoost algorithm accurately predicts TiPN using 18 clinical features and 14 genetic variables. With the ability to identify high-risk patients using single nucleotide polymorphisms, it offers a feasible option for improving thalidomide efficacy in CD patients.

Quantitative assessment of retinal microvascular remodeling in eyes that underwent idiopathic epiretinal membrane surgery

Purpose: To explore the surgical outcomes of the macular microvasculature and visual function in eyes with idiopathic epiretinal membrane (iERM) using spectral-domain optical coherence tomography angiography (SD-OCTA). Methods: This observational, cross-sectional study included 41 participants who underwent iERM surgery with a 3-month (3M) follow-up. Forty-one healthy eyes formed the control group. The assessments included best-corrected visual acuity (BCVA) and mean sensitivity (MS) by microperimetry and SD-OCTA assessment of vessel tortuosity (VT), vessel density (VD), foveal avascular zone, and retinal thickness (RT). Results: The findings showed statistically significant differences in VT, foveal avascular zone parameters, RT, BCVA, and MS between the iERM and control groups (p < 0.05). After iERM surgery, the macular VT, SCP VD, and RT decreased significantly (p < 0.01) while the DCP VD increased (p = 0.029). The BCVA improved significantly (p < 0.001) and was associated with the MS (rs = -0.377, p = 0.015). MS was associated with the SCP VD and RT at 3M (SCP VD rs = 0.511, p = 0.001; RT rs = 0.456, p = 0.003). In the superior quadrant, the MS improved significantly (p < 0.001) and the improvement of MS was associated with the reduction of VT (β = -0.330, p = 0.034). Conclusion: Microcirculatory remodeling and perfusion recovery were observed within 3 months after iERM surgery. VT was a novel index for evaluating the morphology of the retinal microvasculature in eyes with iERM and was associated with MS in the superior quadrant.

Construction of machine learning tools to predict threatened miscarriage in the first trimester based on AEA, progesterone and β-hCG in China: a multicentre, observational, case-control study

Background

Endocannabinoid anandamide (AEA), progesterone (P4) and β-human chorionic gonadotrophin (β-hCG) are associated with the threatened miscarriage in the early stage. However, no study has investigated whether combing these three hormones could predict threatened miscarriage. Thus, we aim to establish machine learning models utilizing these three hormones to predict threatened miscarriage risk.

Methods

This is a multicentre, observational, case-control study involving 215 pregnant women. We recruited 119 normal pregnant women and 96 threatened miscarriage pregnant women including 58 women with ongoing pregnancy and 38 women with inevitable miscarriage. P4 and β-hCG levels were detected by chemiluminescence immunoassay assay. The level of AEA was tested by ultra-high-performance liquid chromatography-tandem mass spectrometry. Six predictive machine learning models were established and evaluated by the confusion matrix, area under the receiver operating characteristic (ROC) curve (AUC), accuracy and precision.

Results

The median concentration of AEA was significantly lower in the healthy pregnant women group than that in the threatened miscarriage group, while the median concentration of P4 was significantly higher in the normal pregnancy group than that in the threatened miscarriage group. Only the median level of P4 was significantly lower in the inevitable miscarriage group than that in the ongoing pregnancy group. Moreover, AEA is strongly positively correlated with threatened miscarriage, while P4 is negatively correlated with both threatened miscarriage and inevitable miscarriage. Interestingly, AEA and P4 are negatively correlated with each other. Among six models, logistic regression (LR), support vector machine (SVM) and multilayer perceptron (MLP) models obtained the AUC values of 0.75, 0.70 and 0.70, respectively; and their accuracy and precision were all above 0.60. Among these three models, the LR model showed the highest accuracy (0.65) and precision (0.70) to predict threatened miscarriage.

Conclusions

The LR model showed the highest overall predictive power, thus machine learning combined with the level of AEA, P4 and β-hCG might be a new approach to predict the threatened miscarriage risk in the near feature.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12884-022-05025-y.

Analysis of COVID-19 severity from the perspective of coagulation index using evolutionary machine learning with enhanced brain storm optimization

Coronavirus 2019 (COVID-19) is an extreme acute respiratory syndrome. Early diagnosis and accurate assessment of COVID-19 are not available, resulting in ineffective therapeutic therapy. This study designs an effective intelligence framework to early recognition and discrimination of COVID-19 severity from the perspective of coagulation indexes. The framework is proposed by integrating an enhanced new stochastic optimizer, a brain storm optimizing algorithm (EBSO), with an evolutionary machine learning algorithm called EBSO-SVM. Fast convergence and low risk of the local stagnant can be guaranteed for EBSO with added by Harris hawks optimization (HHO), and its property is verified on 23 benchmarks. Then, the EBSO is utilized to perform parameter optimization and feature selection simultaneously for support vector machine (SVM), and the presented EBSO-SVM early recognition and discrimination of COVID-19 severity in terms of coagulation indexes using COVID-19 clinical data. The classification performance of the EBSO-SVM is very promising, reaching 91.9195% accuracy, 90.529% Matthews correlation coefficient, 90.9912% Sensitivity and 88.5705% Specificity on COVID-19. Compared with other existing state-of-the-art methods, the EBSO-SVM in this paper still shows obvious advantages in multiple metrics. The statistical results demonstrate that the proposed EBSO-SVM shows predictive properties for all metrics and higher stability, which can be treated as a computer-aided technique for analysis of COVID-19 severity from the perspective of coagulation.

A Self-Supervised Model Advance OCTA Image Disease Diagnosis

Due to the lack of medical image datasets, transfer learning/fine-tuning is generally used to realize disease detection (mainly the ImageNet transfer model). Significant differences of dominance between natural and medical images seriously restrict the performance of the model. In this paper, a contrastive learning method (BY-OCTA) combined with patient metadata is proposed to detect the pathology in fundus OCTA images. This method uses the patient’s metadata to construct positive sample pairs. By introducing super-parameters into the loss function, we can reasonably adjust the approximate proportion of the same patient metadata sample pair, so as to produce a better representation and initialization model. This paper evaluates the performance of downstream tasks by fine-tuning the multi-layer perceptron of the model. Experiments show that the linear model pretrained by BY-OCTA is better than that pretrained by ImageNet and BYOL on multiple datasets. Furthermore, in the case of limited labeled training data, BY-OCTA provides the most significant benefit. This shows that the BY-OCTA pretraining model has better characterization extraction ability and transferability. This method allows a flexible combination of medical opinions and uses metadata to construct positive sample pairs, which can be widely used in medical image interpretation.

Machine Learning Based on Diffusion Kurtosis Imaging Histogram Parameters for Glioma Grading

Glioma grading plays an important role in surgical resection. We investigated the ability of different feature reduction methods in support vector machine (SVM)-based diffusion kurtosis imaging (DKI) histogram parameters to distinguish glioma grades. A total of 161 glioma patients who underwent magnetic resonance imaging (MRI) from January 2017 to January 2020 were included retrospectively. The patients were divided into low-grade (n = 61) and high-grade (n = 100) groups. Parametric DKI maps were derived, and 45 features from the DKI maps were extracted semi-automatically for analysis. Three feature selection methods [principal component analysis (PCA), recursive feature elimination (RFE) and least absolute shrinkage and selection operator (LASSO)] were used to establish the glioma grading model with an SVM classifier. To evaluate the performance of SVM models, the receiver operating characteristic (ROC) curves of SVM models for distinguishing glioma grades were compared with those of conventional statistical methods. The conventional ROC analysis showed that mean diffusivity (MD) variance, MD skewness and mean kurtosis (MK) C50 could effectively distinguish glioma grades, particularly MD variance. The highest classification distinguishing AUC was found using LASSO at 0.904 ± 0.069. In comparison, classification AUC by PCA was 0.866 ± 0.061, and 0.899 ± 0.079 by RFE. The SVM-PCA model with the lowest AUC among the SVM models was significantly better than the conventional ROC analysis (z = 1.947, p = 0.013). These findings demonstrate the superiority of DKI histogram parameters by LASSO analysis and SVM for distinguishing glioma grades.

ADC-Net: An Open-Source Deep Learning Network for Automated Dispersion Compensation in Optical Coherence Tomography

Chromatic dispersion is a common problem to degrade the system resolution in optical coherence tomography (OCT). This study is to develop a deep learning network for automated dispersion compensation (ADC-Net) in OCT. The ADC-Net is based on a modified UNet architecture which employs an encoder-decoder pipeline. The input section encompasses partially compensated OCT B-scans with individual retinal layers optimized. Corresponding output is a fully compensated OCT B-scan with all retinal layers optimized. Two numeric parameters, i.e., peak signal to noise ratio (PSNR) and structural similarity index metric computed at multiple scales (MS-SSIM), were used for objective assessment of the ADC-Net performance and optimal values of 29.95 ± 2.52 dB and 0.97 ± 0.014 were obtained respectively. Comparative analysis of training models, including single, three, five, seven and nine input channels were implemented. The mode with five-input channels was observed to be optimal for ADC-Net training to achieve robust dispersion compensation in OCT.

Deep Learning Image Analysis of Optical Coherence Tomography Angiography Measured Vessel Density Improves Classification of Healthy and Glaucoma Eyes

PURPOSE

To compare convolutional neural network (CNN) analysis of en face vessel density images to gradient boosting classifier (GBC) analysis of instrument-provided, feature-based optical coherence tomography angiography (OCTA) vessel density measurements and OCT retinal nerve fiber layer (RNFL) thickness measurements for classifying healthy and glaucomatous eyes.

DESIGN

Comparison of diagnostic approaches.

METHODS

A total of 130 eyes of 80 healthy individuals and 275 eyes of 185 glaucoma patients with optic nerve head (ONH) OCTA and OCT imaging were included. Classification performance of a VGG16 CNN trained and tested on entire en face 4.5 × 4.5-mm radial peripapillary capillary OCTA ONH images was compared to the performance of separate GBC models trained and tested on standard OCTA and OCT measurements. Five-fold cross-validation was used to test predictions for CNNs and GBCs. Areas under the precision recall curves (AUPRC) were calculated to control for training/test set size imbalance and were compared.

RESULTS

Adjusted AUPRCs for GBC models were 0.89 (95% CI = 0.82, 0.92) for whole image vessel density GBC, 0.89 (0.83, 0.92) for whole image capillary density GBC, 0.91 (0.88, 0.93) for combined whole image vessel and whole image capillary density GBC, and 0.93 (0.91, 095) for RNFL thickness GBC. The adjusted AUPRC using CNN analysis of en face vessel density images was 0.97 (0.95, 0.99) resulting in significantly improved classification compared to GBC OCTA-based results and GBC OCT-based results (P ≤ 0.01 for all comparisons).

CONCLUSION

Deep learning en face image analysis improves on feature-based GBC models for classifying healthy and glaucoma eyes.

Artificial intelligence in OCT angiography

Optical coherence tomographic angiography (OCTA) is a non-invasive imaging modality that provides three-dimensional, information-rich vascular images. With numerous studies demonstrating unique capabilities in biomarker quantification, diagnosis, and monitoring, OCTA technology has seen rapid adoption in research and clinical settings. The value of OCTA imaging is significantly enhanced by image analysis tools that provide rapid and accurate quantification of vascular features and pathology. Today, the most powerful image analysis methods are based on artificial intelligence (AI). While AI encompasses a large variety of techniques, machine-learning-based, and especially deep-learning-based, image analysis provides accurate measurements in a variety of contexts, including different diseases and regions of the eye. Here, we discuss the principles of both OCTA and AI that make their combination capable of answering new questions. We also review contemporary applications of AI in OCTA, which include accurate detection of pathologies such as choroidal neovascularization, precise quantification of retinal perfusion, and reliable disease diagnosis.

Evolutionary warning system for COVID-19 severity: Colony predation algorithm enhanced extreme learning machine

Coronavirus Disease 2019 (COVID-19) was distributed globally at the end of December 2019 due to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Early diagnosis and successful COVID-19 assessment are missing, clinical care is ineffective, and deaths are high. In this study, we investigate whether the level of biochemical indicators helps to discriminate and classify the severity of the COVID-19 using the machine learning method. This research creates an efficient intelligence method for the diagnosis of COVID-19 from the perspective of biochemical indexes. The framework is proposed by integrating an enhanced new stochastic called the colony predation algorithm (CPA) with a kernel extreme learning machine (KELM), abbreviated as ECPA-KELM. The core feature of the approach is the ECPA algorithm which incorporates the two main operators that have been abstained from the grey wolf optimizer and moth-flame optimizer to improve and restore the CPA research functions and are simultaneously used to optimize the parameters and to select features for KELM. The ECPA output is checked thoroughly using IEEE CEC2017 benchmark to verify the capacity of the proposed methodology. Finally, in the diagnosis of COVID-19 using biochemical indexes, the designed ECPA-KELM model and other competing KELM models based on other optimization are used. Checking statistical results will display improved predictive properties for all metrics and higher stability. ECPA-KELM can also be used to discriminate and classify the severity of the COVID-19 as a possible computer-aided method and provide effective early warning for the treatment and diagnosis of COVID-19.

Machine learning in optical coherence tomography angiography

Optical coherence tomography angiography (OCTA) offers a noninvasive label-free solution for imaging retinal vasculatures at the capillary level resolution. In principle, improved resolution implies a better chance to reveal subtle microvascular distortions associated with eye diseases that are asymptomatic in early stages. However, massive screening requires experienced clinicians to manually examine retinal images, which may result in human error and hinder objective screening. Recently, quantitative OCTA features have been developed to standardize and document retinal vascular changes. The feasibility of using quantitative OCTA features for machine learning classification of different retinopathies has been demonstrated. Deep learning-based applications have also been explored for automatic OCTA image analysis and disease classification. In this article, we summarize recent developments of quantitative OCTA features, machine learning image analysis, and classification.

The Role of Telemedicine, In-Home Testing and Artificial Intelligence to Alleviate an Increasingly Burdened Healthcare System: Diabetic Retinopathy

In the presence of the ever-increasing incidence of diabetes mellitus (DM), the prevalence of diabetic eye disease (DED) is also growing. Despite many improvements in diabetic care, DM remains a leading cause of visual impairment in working-age patients. So far, prevention has been the best way to protect vision. The sooner we diagnose DED, the more effective the treatment is. Thus, diabetic retinopathy (DR) screening, especially with imaging techniques, is a method of choice for vision protection. To alleviate the burden of diabetic patients who need ophthalmic care, telemedicine and in-home testing are used, supported by artificial intelligence (AI) algorithms. This is why we decided to evaluate current image teleophthalmology methods used for DR screening. We searched the PubMed platform for papers published over the last 5 years (2015–2020) using the following key words: telemedicine in diabetic retinopathy screening, diabetic retinopathy screening, automated diabetic retinopathy screening, artificial intelligence in diabetic retinopathy screening, smartphone diabetic retinopathy testing. We have included 118 original articles meeting the above criteria, discussing imaging diabetic retinopathy screening methods. We have found that fundus cameras, stable or mobile, are most commonly used for retinal photography, with portable fundus cameras also relatively common. Other possibilities involve the use of ultra-wide-field (UWF) imaging and even optical coherence tomography (OCT) devices for DR screening. Also, the role of smartphones is increasingly recognized in the field. Retinal fundus images are assessed by humans instantly or remotely, while AI algorithms seem to be useful tools facilitating retinal image assessment. The common use of smartphones and availability of relatively cheap, easy-to-use adapters for retinal photographs augmented by AI algorithms make it possible for eye fundus photographs to be taken by non-specialists and in non-medical setting. This opens the way for in-home testing conducted on a much larger scale in the future. In conclusion, based on current DR screening techniques, we can suggest that the future practice of eye care specialists will be widely supported by AI algorithms, and this way will be more effective.

Improved differentiation classification of variable precision artificial intelligence higher education management

The wide application of artificial intelligence technology in various fields has accelerated the pace of people exploring the hidden information behind large amounts of data. People hope to use data mining methods to conduct effective research on higher education management, and decision tree classification algorithm as a data analysis method in data mining technology, high-precision classification accuracy, intuitive decision results, and high generalization ability make it become a more ideal method of higher education management. Aiming at the sensitivity of data processing and decision tree classification to noisy data, this paper proposes corresponding improvements, and proposes a variable precision rough set attribute selection standard based on scale function, which considers both the weighted approximation accuracy and attribute value of the attribute. The number improves the anti-interference ability of noise data, reduces the bias in attribute selection, and improves the classification accuracy. At the same time, the suppression factor threshold, support and confidence are introduced in the tree pre-pruning process, which simplifies the tree structure. The comparative experiments on standard data sets show that the improved algorithm proposed in this paper is better than other decision tree algorithms and can effectively realize the differentiated classification of higher education management.

Artificial Intelligence (AI) and Retinal Optical Coherence Tomography (OCT)

Ophthalmology has been at the forefront of medical specialties adopting artificial intelligence. This is primarily due to the "image-centric" nature of the field. Thanks to the abundance of patients' OCT scans, analysis of OCT imaging has greatly benefited from artificial intelligence to expand patient screening and facilitate clinical decision-making.In this review, we define the concepts of artificial intelligence, machine learning, and deep learning and how different artificial intelligence algorithms have been applied in OCT image analysis for disease screening, diagnosis, management, and prognosis.Finally, we address some of the challenges and limitations that might affect the incorporation of artificial intelligence in ophthalmology. These limitations mainly revolve around the quality and accuracy of datasets used in the algorithms and their generalizability, false negatives, and the cultural challenges around the adoption of the technology.

Evaluation of Accuracy and Agreement of Optical Coherence Tomography Angiography Interpretation of Common Retinal Findings and Diagnoses

PURPOSE

To evaluate the accuracy and agreement of optical coherence tomography angiography (OCTA) interpretation in cases of common retinal findings and diagnoses, and to evaluate the effect of OCT B-scans on OCTA interpretations.

METHODS

This is a case series consisting of a questionnaire with 8 cases demonstrating common retinal conditions of normal, age-related macular degeneration (AMD) and diabetic retinopathy (DR). Each case included OCTA images, and 58 participants were asked to identify retinal findings and provide a diagnosis. Following OCTA interpretation, the corresponding OCT B-scans were revealed and the participants were asked again to identify retinal findings and provide a diagnosis. The rates of accuracy and agreement for each condition were analyzed.

RESULTS

Overall the rates of accurate diagnosis and identification of retinal findings were 37.4 and 61.6%, respectively. Following addition of the OCT B-scans, the rates increased to 61.6 and 79.4%, respectively (p < 0.001 for both). A significant improvement in correct interpretation occurred in the normal and AMD cases, but not in the DR cases. There was no correlation with length of experience or self-reported familiarity with OCTA.

DISCUSSION

Considerable variability exists in OCTA interpretation, with mediocre rates of accuracy and agreement between clinicians. Increased familiarity as well as future automation advances will be needed to improve OCTA interpretation accuracy and uniformity.

Noninvasive temporal detection of early retinal vascular changes during diabetes

Diabetes associated complications, including diabetic retinopathy and loss of vision, are major health concerns. Detecting early retinal vascular changes during diabetes is not well documented, and only few studies have addressed this domain. The purpose of this study was to noninvasively evaluate temporal changes in retinal vasculature at very early stages of diabetes using fundus images from preclinical models of diabetes. Non-diabetic and Akita/+ male mice with different duration of diabetes were subjected to fundus imaging using a Micron III imaging system. The images were obtained from 4 weeks- (onset of diabetes), 8 weeks-, 16 weeks-, and 24 weeks-old male Akita/+ and non-diabetic mice. In total 104 fundus images were subjected to analysis for various feature extractions. A combination of Canny Edge Detector and Angiogenesis Analyzer plug-ins in ImageJ were utilized to quantify various retinal vascular changes in fundus images. Statistical analyses were conducted to determine significant differences in the various extracted features from fundus images of diabetic and non-diabetic animals. Our novel image analysis method led to extraction of over 20 features. These results indicated that some of these features were significantly changed with a short duration of diabetes, and others remained the same but changed after longer duration of diabetes. These patterns likely distinguish acute (protective) and chronic (damaging) associated changes with diabetes. We show that with a combination of various plugging one can extract over 20 features from retinal vasculature fundus images. These features change during diabetes, thus allowing the quantification of quality of retinal vascular architecture as biomarkers for disease progression. In addition, our method was able to identify unique differences among diabetic mice with different duration of diabetes. The ability to noninvasively detect temporal retinal vascular changes during diabetes could lead to identification of specific markers important in the development and progression of diabetes mediated-microvascular changes, evaluation of therapeutic interventions, and eventual reversal of these changes in order to stop or delay disease progression.

Classification of advanced and early stages of diabetic retinopathy from non-diabetic subjects by an ordinary least squares modeling method applied to OCTA images

As the prevalence of diabetic retinopathy (DR) continues to rise, there is a need to develop computer-aided screening methods. The current study reports and validates an ordinary least squares (OLS) method to model optical coherence tomography angiography (OCTA) images and derive OLS parameters for classifying proliferative DR (PDR) and no/mild non-proliferative DR (NPDR) from non-diabetic subjects. OLS parameters were correlated with vessel metrics quantified from OCTA images and were used to determine predicted probabilities of PDR, no/mild NPDR, and non-diabetics. The classification rates of PDR and no/mild NPDR from non-diabetic subjects were 94% and 91%, respectively. The method had excellent predictive ability and was validated. With further development, the method may have potential clinical utility and contribute to image-based computer-aided screening and classification of stages of DR and other ocular and systemic diseases.

Transfer Learning for Automated OCTA Detection of Diabetic Retinopathy

Purpose

To test the feasibility of using deep learning for optical coherence tomography angiography (OCTA) detection of diabetic retinopathy.

Methods

A deep-learning convolutional neural network (CNN) architecture, VGG16, was employed for this study. A transfer learning process was implemented to retrain the CNN for robust OCTA classification. One dataset, consisting of images of 32 healthy eyes, 75 eyes with diabetic retinopathy (DR), and 24 eyes with diabetes but no DR (NoDR), was used for training and cross-validation. A second dataset consisting of 20 NoDR and 26 DR eyes was used for external validation. To demonstrate the feasibility of using artificial intelligence (AI) screening of DR in clinical environments, the CNN was incorporated into a graphical user interface (GUI) platform.

Results

With the last nine layers retrained, the CNN architecture achieved the best performance for automated OCTA classification. The cross-validation accuracy of the retrained classifier for differentiating among healthy, NoDR, and DR eyes was 87.27%, with 83.76% sensitivity and 90.82% specificity. The AUC metrics for binary classification of healthy, NoDR, and DR eyes were 0.97, 0.98, and 0.97, respectively. The GUI platform enabled easy validation of the method for AI screening of DR in a clinical environment.

Conclusions

With a transfer learning process for retraining, a CNN can be used for robust OCTA classification of healthy, NoDR, and DR eyes. The AI-based OCTA classification platform may provide a practical solution to reducing the burden of experienced ophthalmologists with regard to mass screening of DR patients.

Translational Relevance

Deep-learning-based OCTA classification can alleviate the need for manual graders and improve DR screening efficiency.

Diving Deep into Deep Learning: An Update on Artificial Intelligence in Retina

Purpose of Review

In the present article, we will provide an understanding and review of artificial intelligence in the subspecialty of retina and its potential applications within the specialty.

Recent Findings

Given the significant use of diagnostic imaging within retina, this subspecialty is a fitting area for the incorporation of artificial intelligence. Researchers have aimed at creating models to assist in the diagnosis and management of retinal disease as well as in the prediction of disease course and treatment response. Most of this work thus far has focused on diabetic retinopathy, age-related macular degeneration, and retinopathy of prematurity, although other retinal diseases have started to be explored as well.

Summary

Artificial intelligence is well-suited to transform the practice of ophthalmology. A basic understanding of the technology is important for its effective implementation and growth.

Comparison and development of machine learning tools for the prediction of chronic obstructive pulmonary disease in the Chinese population

Background

Chronic obstructive pulmonary disease (COPD) is a major public health problem and cause of mortality worldwide. However, COPD in the early stage is usually not recognized and diagnosed. It is necessary to establish a risk model to predict COPD development.

Methods

A total of 441 COPD patients and 192 control subjects were recruited, and 101 single-nucleotide polymorphisms (SNPs) were determined using the MassArray assay. With 5 clinical features as well as SNPs, 6 predictive models were established and evaluated in the training set and test set by the confusion matrix AU-ROC, AU-PRC, sensitivity (recall), specificity, accuracy, F1 score, MCC, PPV (precision) and NPV. The selected features were ranked.

Results

Nine SNPs were significantly associated with COPD. Among them, 6 SNPs (rs1007052, OR = 1.671, P = 0.010; rs2910164, OR = 1.416, P < 0.037; rs473892, OR = 1.473, P < 0.044; rs161976, OR = 1.594, P < 0.044; rs159497, OR = 1.445, P < 0.045; and rs9296092, OR = 1.832, P < 0.045) were risk factors for COPD, while 3 SNPs (rs8192288, OR = 0.593, P < 0.015; rs20541, OR = 0.669, P < 0.018; and rs12922394, OR = 0.651, P < 0.022) were protective factors for COPD development. In the training set, KNN, LR, SVM, DT and XGboost obtained AU-ROC values above 0.82 and AU-PRC values above 0.92. Among these models, XGboost obtained the highest AU-ROC (0.94), AU-PRC (0.97), accuracy (0.91), precision (0.95), F1 score (0.94), MCC (0.77) and specificity (0.85), while MLP obtained the highest sensitivity (recall) (0.99) and NPV (0.87). In the validation set, KNN, LR and XGboost obtained AU-ROC and AU-PRC values above 0.80 and 0.85, respectively. KNN had the highest precision (0.82), both KNN and LR obtained the same highest accuracy (0.81), and KNN and LR had the same highest F1 score (0.86). Both DT and MLP obtained sensitivity (recall) and NPV values above 0.94 and 0.84, respectively. In the feature importance analyses, we identified that AQCI, age, and BMI had the greatest impact on the predictive abilities of the models, while SNPs, sex and smoking were less important.

Conclusions

The KNN, LR and XGboost models showed excellent overall predictive power, and the use of machine learning tools combining both clinical and SNP features was suitable for predicting the risk of COPD development.

Quantitative optical coherence tomography angiography: A review

As a new optical coherence tomography (OCT) modality, OCT angiography (OCTA) provides a noninvasive method to detect microvascular distortions correlated with eye conditions. By providing unparalleled capability to differentiate individual plexus layers in the retina, OCTA has demonstrated its excellence in clinical management of diabetic retinopathy, glaucoma, sickle cell retinopathy, diabetic macular edema, and other eye diseases. Quantitative OCTA analysis of retinal and choroidal vasculatures is essential to standardize objective interpretations of clinical outcome. Quantitative features, including blood vessel tortuosity, blood vessel caliber, blood vessel density, vessel perimeter index, fovea avascular zone area, fovea avascular zone contour irregularity, vessel branching coefficient, vessel branching angle, branching width ratio, and choroidal vascular analysis have been established for objective OCTA assessment. Moreover, differential artery–vein analysis has been recently demonstrated to improve OCTA performance for objective detection and classification of eye diseases. In this review, technical rationales and clinical applications of these quantitative OCTA features are summarized, and future prospects for using these quantitative OCTA features for artificial intelligence classification of eye conditions are discussed.

OCT Angiography Assessment of Retinal Microvascular Changes in Diabetic Eyes in an Urban Safety-Net Hospital

PURPOSE

To determine whether quantitative OCT angiography (OCTA) parameters can be used to distinguish among eyes at various stages of diabetic retinopathy (DR) in an urban safety-net hospital population.

DESIGN

Prospective cross-sectional study.

PARTICIPANTS

Three hundred twenty-nine eyes from 329 patients were included in this study: 90 nondiabetic patients, 170 diabetic patients without DR, 57 diabetic patients with mild to moderate nonproliferative DR (NPDR), and 12 diabetic patients with severe NPDR to proliferative DR.

METHODS

Patients underwent OCTA imaging and ultra-widefield fundus photography at Zuckerberg San Francisco General Hospital and Trauma Center between April and October 2018. For participants with diabetes, imaging was classified according to DR severity by a telemedicine reading center. Eight OCTA parameters were analyzed. Perfusion density and vessel length density (VD) were examined from both the superficial capillary plexus (SCP) and deep capillary plexus. The other 4 parameters were examined only from the SCP. Total extrafoveal avascular area (tEAA) was based on the area of absent capillary vessels. Foveal avascular zone (FAZ)-related metrics consisted of FAZ area, FAZ circularity index, and FAZ acircularity index.

MAIN OUTCOME MEASURES

Area under the receiver operating characteristic curve (AUC) for OCTA parameters to distinguish among groups according to DR severity.

RESULTS

All OCTA parameters demonstrated a significant relationship with DR severity (P < 0.05). No significant difference was found when comparing nondiabetic participants versus diabetic participants without retinopathy. The FAZ area was the only metric that demonstrated a significant difference between genders: mean of 0.29±0.12 mm² in men and 0.34±0.13 mm² in women (P < 0.001). Receiver operating characteristic curve analyses showed that tEAA had the highest AUC when comparing various stages of the disease.

CONCLUSIONS

In this urban, public hospital population, quantification of retinal vascular findings with OCTA imaging was a useful means of distinguishing patients according to DR severity. Because these results were similar to those of other tertiary referral centers, it would be reasonable to perform further DR-related OCTA studies in this population and expect generalizable results.

Insights into Amyotrophic Lateral Sclerosis from a Machine Learning Perspective

Objective: Amyotrophic lateral sclerosis (ALS) disease state prediction usually assumes linear progression and uses a classifier evaluated by its accuracy. Since disease progression is not linear, and the accuracy measurement cannot tell large from small prediction errors, we dispense with the linearity assumption and apply ordinal classification that accounts for error severity. In addition, we identify the most influential variables in predicting and explaining the disease. Furthermore, in contrast to conventional modeling of the patient’s total functionality, we also model separate patient functionalities (e.g., in walking or speaking). Methods: Using data from 3772 patients from the Pooled Resource Open-Access ALS Clinical Trials (PRO-ACT) database, we introduce and train ordinal classifiers to predict patients’ disease state in their last clinic visit, while accounting differently for different error severities. We use feature-selection methods and the classifiers themselves to determine the most influential variables in predicting the disease from demographic, clinical, and laboratory data collected in either the first, last, or both clinic visits, and the Bayesian network classifier to identify interrelations among these variables and their relations with the disease state. We apply these methods to model each of the patient functionalities. Results: We show the error distribution in ALS state prediction and demonstrate that ordinal classifiers outperform classifiers that do not account for error severity. We identify clinical and lab test variables influential to prediction of different ALS functionalities and their interrelations, and specific value combinations of these variables that occur more frequently in patients with severe deterioration than in patients with mild deterioration and vice versa. Conclusions: Ordinal classification of ALS state is superior to conventional classification. Identification of influential ALS variables and their interrelations help explain disease mechanism. Modeling of patient functionalities separately allows relation of variables and their connections to different aspects of the disease as may be expressed in different body segments.

Aiding the Diagnosis of Diabetic and Hypertensive Retinopathy Using Artificial Intelligence-Based Semantic Segmentation

Automatic segmentation of retinal images is an important task in computer-assisted medical image analysis for the diagnosis of diseases such as hypertension, diabetic and hypertensive retinopathy, and arteriosclerosis. Among the diseases, diabetic retinopathy, which is the leading cause of vision detachment, can be diagnosed early through the detection of retinal vessels. The manual detection of these retinal vessels is a time-consuming process that can be automated with the help of artificial intelligence with deep learning. The detection of vessels is difficult due to intensity variation and noise from non-ideal imaging. Although there are deep learning approaches for vessel segmentation, these methods require many trainable parameters, which increase the network complexity. To address these issues, this paper presents a dual-residual-stream-based vessel segmentation network (Vess-Net), which is not as deep as conventional semantic segmentation networks, but provides good segmentation with few trainable parameters and layers. The method takes advantage of artificial intelligence for semantic segmentation to aid the diagnosis of retinopathy. To evaluate the proposed Vess-Net method, experiments were conducted with three publicly available datasets for vessel segmentation: digital retinal images for vessel extraction (DRIVE), the Child Heart Health Study in England (CHASE-DB1), and structured analysis of retina (STARE). Experimental results show that Vess-Net achieved superior performance for all datasets with sensitivity (Se), specificity (Sp), area under the curve (AUC), and accuracy (Acc) of 80.22%, 98.1%, 98.2%, and 96.55% for DRVIE; 82.06%, 98.41%, 98.0%, and 97.26% for CHASE-DB1; and 85.26%, 97.91%, 98.83%, and 96.97% for STARE dataset.