Enhancing the Accuracy of Glaucoma Detection from OCT Probability Maps using Convolutional Neural Networks

Recognition of eye diseases based on deep neural networks for transfer learning and improved D-S evidence theory

BACKGROUND

Human vision has inspired significant advancements in computer vision, yet the human eye is prone to various silent eye diseases. With the advent of deep learning, computer vision for detecting human eye diseases has gained prominence, but most studies have focused only on a limited number of eye diseases.

RESULTS

Our model demonstrated a reduction in inherent bias and enhanced robustness. The fused network achieved an Accuracy of 0.9237, Kappa of 0.878, F1 Score of 0.914 (95% CI [0.875-0.954]), Precision of 0.945 (95% CI [0.928-0.963]), Recall of 0.89 (95% CI [0.821-0.958]), and an AUC value of ROC at 0.987. These metrics are notably higher than those of comparable studies.

CONCLUSIONS

Our deep neural network-based model exhibited improvements in eye disease recognition metrics over models from peer research, highlighting its potential application in this field.

METHODS

In deep learning-based eye recognition, to improve the learning efficiency of the model, we train and fine-tune the network by transfer learning. In order to eliminate the decision bias of the models and improve the credibility of the decisions, we propose a model decision fusion method based on the D-S theory. However, D-S theory is an incomplete and conflicting theory, we improve and eliminate the existed paradoxes, propose the improved D-S evidence theory(ID-SET), and apply it to the decision fusion of eye disease recognition models.

Phase Difference Network for Efficient Differentiation of Hepatic Tumors with Multi-Phase CT

Liver cancer has been one of the top causes of cancer-related death. For developing an accurate treatment strategy and raising the survival rate, the differentiation of liver cancers is essential. Multiphase CT recently acts as the primary examination method for clinical diagnosis. Deep learning techniques based on multiphase CT have been proposed to distinguish hepatic cancers. However, due to the recurrent mechanism, RNN-based approaches require expensive calculations whereas CNN-based models fail to explicitly establish temporal correlations among phases. In this paper, we proposed a phase difference network, termed as Phase Difference Network (PDN), to identify two liver cancer, hepatocellular carcinoma and intrahepatic cholangiocarcinoma, from four-phase CT. Specifically, the phase difference was used as interphase temporal information in a differential attention module, which enhanced the feature representation. Additionally, utilizing a multihead self-attention module, a transformer-based classification module was employed to explore the long-term context and capture the temporal relation between phases. Clinical datasets are used in experiments to compare the performance of the proposed strategy versus conventional approaches. The results indicate that the proposed method outperforms the traditional deep learning based methods.

Automatic Diagnosis of Glaucoma from Retinal Images Using Deep Learning Approach

Glaucoma is characterized by increased intraocular pressure and damage to the optic nerve, which may result in irreversible blindness. The drastic effects of this disease can be avoided if it is detected at an early stage. However, the condition is frequently detected at an advanced stage in the elderly population. Therefore, early-stage detection may save patients from irreversible vision loss. The manual assessment of glaucoma by ophthalmologists includes various skill-oriented, costly, and time-consuming methods. Several techniques are in experimental stages to detect early-stage glaucoma, but a definite diagnostic technique remains elusive. We present an automatic method based on deep learning that can detect early-stage glaucoma with very high accuracy. The detection technique involves the identification of patterns from the retinal images that are often overlooked by clinicians. The proposed approach uses the gray channels of fundus images and applies the data augmentation technique to create a large dataset of versatile fundus images to train the convolutional neural network model. Using the ResNet-50 architecture, the proposed approach achieved excellent results for detecting glaucoma on the G1020, RIM-ONE, ORIGA, and DRISHTI-GS datasets. We obtained a detection accuracy of 98.48%, a sensitivity of 99.30%, a specificity of 96.52%, an AUC of 97%, and an F1-score of 98% by using the proposed model on the G1020 dataset. The proposed model may help clinicians to diagnose early-stage glaucoma with very high accuracy for timely interventions.

Performance evaluation of various deep learning based models for effective glaucoma evaluation using optical coherence tomography images

Glaucoma is the dominant reason for irreversible blindness worldwide, and its best remedy is early and timely detection. Optical coherence tomography has come to be the most commonly used imaging modality in detecting glaucomatous damage in recent years. Deep Learning using Optical Coherence Tomography Modality helps in predicting glaucoma more accurately and less tediously. This experimental study aims to perform glaucoma prediction using eight different ImageNet models from Optical Coherence Tomography of Glaucoma. A thorough investigation is performed to evaluate these models' performances on various efficiency metrics, which will help discover the best performing model. Every net is tested on three different optimizers, namely Adam, Root Mean Squared Propagation, and Stochastic Gradient Descent, to find the best relevant results. An attempt has been made to improvise the performance of models using transfer learning and fine-tuning. The work presented in this study was initially trained and tested on a private database that consists of 4220 images (2110 normal optical coherence tomography and 2110 glaucoma optical coherence tomography). Based on the results, the four best-performing models are shortlisted. Later, these models are tested on the well-recognized standard public Mendeley dataset. Experimental results illustrate that VGG16 using the Root Mean Squared Propagation Optimizer attains auspicious performance with 95.68% accuracy. The proposed work concludes that different ImageNet models are a good alternative as a computer-based automatic glaucoma screening system. This fully automated system has a lot of potential to tell the difference between normal Optical Coherence Tomography and glaucomatous Optical Coherence Tomography automatically. The proposed system helps in efficiently detecting this retinal infection in suspected patients for better diagnosis to avoid vision loss and also decreases senior ophthalmologists' (experts) precious time and involvement.

The OCT RNFL Probability Map and Artifacts Resembling Glaucomatous Damage

Purpose

The purpose of this study was to improve the diagnostic ability of the optical coherence tomography (OCT) retinal nerve fiber layer (RNFL) probability (p-) map by understanding the frequency and pattern of artifacts seen on the p-maps of healthy control (HC) eyes resembling glaucomatous damage.

Methods

RNFL p-maps were generated from wide-field OCT cube scans of 2 groups of HC eyes, 200 from a commercial normative group (HC-norm) and 54 from a prospective study group, as well as from 62 patient eyes, which included 32 with early glaucoma (EG). These 32 EG eyes had 24-2 mean deviation (MD) better than -6 dB and perimetric glaucoma as defined by 24-2 and 10-2 criteria. For the HC groups, "glaucoma-like" arcuates were defined as any red region near the temporal half of the disc.

Results

Seven percent of the 200 HC-norm and 11% of the 54 HC RNFL p-maps satisfied the definition of "glaucoma-like," as did all the patients' p-maps. The HC p-maps showed two general patterns of abnormal regions, "arcuate" and "temporal quadrant," and these patterns resembled those seen on some of the RNFL p-maps of the EG eyes. A "vertical midline" rule, which required the abnormal region to cross the vertical midline through the fovea, had a specificity of >99%, and a sensitivity of 75% for EG and 93% for moderate to advanced eyes.

Conclusions

Glaucoma-like artifacts on RNFL p-maps are relatively common and can masquerade as arcuate and/or widespread/temporal damage.

Translational Relevance

A vertical midline rule had excellent specificity. However, other OCT information is necessary to obtain high sensitivity, especially in eyes with early glaucoma.

Natural Disasters Intensity Analysis and Classification Based on Multispectral Images Using Multi-Layered Deep Convolutional Neural Network

Natural disasters not only disturb the human ecological system but also destroy the properties and critical infrastructures of human societies and even lead to permanent change in the ecosystem. Disaster can be caused by naturally occurring events such as earthquakes, cyclones, floods, and wildfires. Many deep learning techniques have been applied by various researchers to detect and classify natural disasters to overcome losses in ecosystems, but detection of natural disasters still faces issues due to the complex and imbalanced structures of images. To tackle this problem, we propose a multilayered deep convolutional neural network. The proposed model works in two blocks: Block-I convolutional neural network (B-I CNN), for detection and occurrence of disasters, and Block-II convolutional neural network (B-II CNN), for classification of natural disaster intensity types with different filters and parameters. The model is tested on 4428 natural images and performance is calculated and expressed as different statistical values: sensitivity (SE), 97.54%; specificity (SP), 98.22%; accuracy rate (AR), 99.92%; precision (PRE), 97.79%; and F1-score (F1), 97.97%. The overall accuracy for the whole model is 99.92%, which is competitive and comparable with state-of-the-art algorithms.

Strategies to Improve Convolutional Neural Network Generalizability and Reference Standards for Glaucoma Detection From OCT Scans

Purpose

To develop and evaluate methods to improve the generalizability of convolutional neural networks (CNNs) trained to detect glaucoma from optical coherence tomography retinal nerve fiber layer probability maps, as well as optical coherence tomography circumpapillary disc (circle) b-scans, and to explore impact of reference standard (RS) on CNN accuracy.

Methods

CNNs previously optimized for glaucoma detection from retinal nerve fiber layer probability maps, and newly developed CNNs adapted for glaucoma detection from optical coherence tomography b-scans, were evaluated on an unseen dataset (i.e., data collected at a different site). Multiple techniques were used to enhance CNN generalizability, including augmenting the training dataset, using multimodal input, and training with confidently rated images. Model performance was evaluated with different RS.

Results

Training with data augmentation and training on confident images enhanced the accuracy of the CNNs for glaucoma detection on a new dataset by 5% to 9%. CNN performance was optimal when a similar RS was used to establish labels both for the training and the testing sets. However, interestingly, the CNNs described here were robust to variation in the RS.

Conclusions

CNN generalizability can be improved with data augmentation, multiple input image modalities, and training on images with confident ratings. CNNs trained and tested with the same RS achieved best accuracy, suggesting that choosing a thorough and consistent RS for training and testing improves generalization to new datasets.

Translational Relevance

Strategies for enhancing CNN generalizability and for choosing optimal RS should be standard practice for CNNs before their deployment for glaucoma detection.

An Evaluation of a New 24-2 Metric for Detecting Early Central Glaucomatous Damage

PURPOSE

We sought to test the hypothesis that a recently proposed pattern standard deviation (PSD) metric, based upon the 24-2 visual field (VF) test, as well as the PSD of the 10-2 VF, will miss central glaucomatous damage confirmed with an objective structure-function method.

DESIGN

Cross-sectional study.

METHODS

A glaucoma (G) group (70 eyes/patients) diagnosed with glaucoma and a 24-2 mean deviation better than -6 dB and a healthy (H) group (45 eyes/patients) had 24-2 and 10-2 VFs and optical coherence tomography (OCT) scans twice within 4 weeks. The PSD(C24-2), based upon the central 12 points of the 24-2, was compared with the PSD(10-2). To evaluate central damage (CD) in G eyes with normal PSD(C24-2) values, a post hoc analysis was combined with a CD reference standard (CD-RS), which was based upon an objective, topographic comparison between abnormal points on the 10-2 VF and OCT probability maps.

RESULTS

The 115 PSD(C24-2) and PSD(10-2) values were significantly correlated (Spearman correclation coefficient: rho = 0.55; P < .001) and the number of G eyes (19) identified as abnormal by the PSD(C24-2) was not significantly different from the number (22) identified by the PSD(10-2) (P = .15). However, based upon the CD-RS, 44 of 70 G eyes were classified as abnormal. The PSD(C24-2) missed 27 (61%) of these 44 eyes, and the PSD(10-2) missed 23 (52%) of these eyes. Post hoc analysis revealed clear CD in most of these eyes.

CONCLUSION

Neither the PSD(C24-2) nor the PSD(10-2) metric is good measure of early CD. Instead we recommend a topographic approach based upon OCT probability maps and a 10-2 VF.

Global optical coherence tomography measures for detecting the progression of glaucoma have fundamental flaws

Objective

To understand the problems involved in using global OCT measures for detecting progression in early glaucoma.

Subjects/Methods

Eyes from 76 patients and 28 healthy controls (HC) had a least two OCT scans at least 1 year apart. To determine the 95% confidence intervals (CI), 151 eyes (49 HC and 102 patients) had at least two scans within 6 months. All eyes had 24-2 mean deviation ≥-6dB. The average (global) thicknesses of the circumpapillary retinal nerve fibre layer (cRNFL), G_ONH, and of the retinal ganglion cell layer plus inner plexiform layer (RGCLP), G_mac, were calculated. Using quantile regression, the 95% CI intervals were determined. Eyes outside the CIs were classified as “progressors.” For a reference standard (RS), four experts evaluated OCT and VF information.

Results

Compared to the RS, 31 of the 76 (40.8%) patient eyes were identified as progressors (RS-P), and 45 patient, and all 28 HC, eyes as nonprogressors (RS-NP). The metrics missed (false negative, FN) 15 (48%) (G_ONH) and 9 (29%) (G_mac) of the 31 RS-P. Further, G_ONH and/or G_mac falsely identified (false positive, FP) 10 (22.2%) of 45 patient RS-NP eyes and 7 (25%) of the 28 HC eyes as progressing. Post-hoc analysis identified three reasons (segmentation, centring, and local damage) for these errors.

Conclusions

Global metrics lead to FPs and FNs because of problems inherent in OCT scanning (segmentation and centring), and to FNs because they can miss local damage. These problems are difficult, if not impossible, to correct, and raise concerns about the advisability of using G_ONH and G_mac for detecting progression.

Robust and Interpretable Convolutional Neural Networks to Detect Glaucoma in Optical Coherence Tomography Images

Recent studies suggest that deep learning systems can now achieve performance on par with medical experts in diagnosis of disease. A prime example is in the field of ophthalmology, where convolutional neural networks (CNNs) have been used to detect retinal and ocular diseases. However, this type of artificial intelligence (AI) has yet to be adopted clinically due to questions regarding robustness of the algorithms to datasets collected at new clinical sites and a lack of explainability of AI-based predictions, especially relative to those of human expert counterparts. In this work, we develop CNN architectures that demonstrate robust detection of glaucoma in optical coherence tomography (OCT) images and test with concept activation vectors (TCAVs) to infer what image concepts CNNs use to generate predictions. Furthermore, we compare TCAV results to eye fixations of clinicians, to identify common decision-making features used by both AI and human experts. We find that employing fine-tuned transfer learning and CNN ensemble learning create end-to-end deep learning models with superior robustness compared to previously reported hybrid deep-learning/machine-learning models, and TCAV/eye-fixation comparison suggests the importance of three OCT report sub-images that are consistent with areas of interest fixated upon by OCT experts to detect glaucoma. The pipeline described here for evaluating CNN robustness and validating interpretable image concepts used by CNNs with eye movements of experts has the potential to help standardize the acceptance of new AI tools for use in the clinic.

An Adoptive Threshold-Based Multi-Level Deep Convolutional Neural Network for Glaucoma Eye Disease Detection and Classification

Glaucoma, an eye disease, occurs due to Retinal damages and it is an ordinary cause of blindness. Most of the available examining procedures are too long and require manual instructions to use them. In this work, we proposed a multi-level deep convolutional neural network (ML-DCNN) architecture on retinal fundus images to diagnose glaucoma. We collected a retinal fundus images database from the local hospital. The fundus images are pre-processed by an adaptive histogram equalizer to reduce the noise of images. The ML-DCNN architecture is used for features extraction and classification into two phases, one for glaucoma detection known as detection-net and the second one is classification-net used for classification of affected retinal glaucoma images into three different categories: Advanced, Moderate and Early. The proposed model is tested on 1338 retinal glaucoma images and performance is measured in the form of different statistical terms known as sensitivity (SE), specificity (SP), accuracy (ACC), and precision (PRE). On average, SE of 97.04%, SP of 98.99%, ACC of 99.39%, and PRC of 98.2% are achieved. The obtained outcomes are comparable to the state-of-the-art systems and achieved competitive results to solve the glaucoma eye disease problems for complex glaucoma eye disease cases.