Detecting Retinal Nerve Fibre Layer Segmentation Errors on Spectral Domain-Optical Coherence Tomography with a Deep Learning Algorithm

Jammal, Alessandro A.; Ogata, Nara G.; Mariottoni, Eduardo B.; Urata, Carla N.; Costa, Vital Paulino; Medeiros, Felipe A.

doi:10.1038/s41598-019-46294-6

Cited by 22 publications

(21 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Class activation maps (CAMs) 20 are a common method where a heat map is generated by projecting the class specific weights of the output classification layer back to the feature maps of the last convolutional layer, thereby highlighting important regions for predicting a particular class. This method has been used in ophthalmic application previously to confirm CNN decision was based off the anterior chamber angle in categorizing angle closure, 21 areas of OCT B-scans associated with various diagnoses 22 , 23 and areas of segmentation error, 24 and area of OCT enface images associated with the diagnosis of glaucoma. 25 There exists several variants of this method that build off of the original CAM paper, 20 including: Grad-Cam, 26 Guided Grad-Cam, 26 Guided Grad-Cam++, 27 and GAIN.…”

Section: Introductionmentioning

confidence: 99%

Ensemble Deep Learning for Diabetic Retinopathy Detection Using Optical Coherence Tomography Angiography

Heisler

Karst

et al. 2020

Trans. Vis. Sci. Tech.

View full text Add to dashboard Cite

To evaluate the role of ensemble learning techniques with deep learning in classifying diabetic retinopathy (DR) in optical coherence tomography angiography (OCTA) images and their corresponding co-registered structural images. Methods: A total of 463 volumes from 380 eyes were acquired using the 3 × 3-mm OCTA protocol on the Zeiss Plex Elite system. Enface images of the superficial and deep capillary plexus were exported from both the optical coherence tomography and OCTA data. Component neural networks were constructed using single data-types and fine-tuned using VGG19, ResNet50, and DenseNet architectures pretrained on ImageNet weights. These networks were then ensembled using majority soft voting and stacking techniques. Results were compared with a classifier using manually engineered features. Class activation maps (CAMs) were created using the original CAM algorithm and Grad-CAM. Results: The networks trained with the VGG19 architecture outperformed the networks trained on deeper architectures. Ensemble networks constructed using the four finetuned VGG19 architectures achieved accuracies of 0.92 and 0.90 for the majority soft voting and stacking methods respectively. Both ensemble methods outperformed the highest single data-type network and the network trained on hand-crafted features. Grad-CAM was shown to more accurately highlight areas of disease. Conclusions: Ensemble learning increases the predictive accuracy of CNNs for classifying referable DR on OCTA datasets. Translational Relevance: Because the diagnostic accuracy of OCTA images is shown to be greater than the manually extracted features currently used in the literature, the proposed methods may be beneficial toward developing clinically valuable solutions for DR diagnoses.

show abstract

Section: Introductionmentioning

confidence: 99%

Ensemble Deep Learning for Diabetic Retinopathy Detection Using Optical Coherence Tomography Angiography

Heisler

Karst

et al. 2020

Trans. Vis. Sci. Tech.

View full text Add to dashboard Cite

show abstract

“…Recently, neural networks and other artificial intelligence (AI) algorithms have been shown to successfully model complex, nonlinear relationships in data from diverse medical fields. 8 – 12 In particular, convolutional neural networks (CNNs) are able to take advantage of spatial information to identify underlying relationships that may not be easily discerned by conventional methods. A few studies have attempted to use AI algorithms to predict visual field results from SDOCT measurements, with good results.…”

Section: Introductionmentioning

confidence: 99%

Artificial Intelligence Mapping of Structure to Function in Glaucoma

Mariottoni

Datta

Dov

et al. 2020

Trans. Vis. Sci. Tech.

Self Cite

View full text Add to dashboard Cite

Purpose:To develop an artificial intelligence (AI)-based structure-function (SF) map relating retinal nerve fiber layer (RNFL) damage on spectral domain optical coherence tomography (SDOCT) to functional loss on standard automated perimetry (SAP). Methods:The study included 26,499 pairs of SAP and SDOCT from 15,173 eyes of 8878 patients with glaucoma or suspected of having the disease extracted from the Duke Glaucoma Registry. The data set was randomly divided at the patient level in training and test sets. A convolutional neural network (CNN) was initially trained and validated to predict the 52 sensitivity threshold points of the 24-2 SAP from the 768 RNFL thickness points of the SDOCT peripapillary scan. Simulated localized RNFL defects of varied locations and depths were created by modifying the normal average peripapillary RNFL profile. The simulated profiles were then fed to the previously trained CNN, and the topographic SF relationships between structural defects and SAP functional losses were investigated. Results:The CNN predictions had an average correlation coefficient of 0.60 (P < 0.001) with the measured values from SAP and a mean absolute error of 4.25 dB. Simulated RNFL defects led to well-defined arcuate or paracentral visual field losses in the opposite hemifield, which varied according to the location and depth of the simulations. Conclusions:A CNN was capable of predicting SAP sensitivity thresholds from SDOCT RNFL thickness measurements and generate an SF map from simulated defects.Translational Relevance: AI-based SF map improves the understanding of how SDOCT losses translate into detectable SAP damage.

show abstract

“…Artefacts and segmentation errors are highly prevalent on SDOCT and result in erroneous estimates of RNFL thickness [5][6][7]11,12 . While Asrani et al identified artefacts in 19.9% of RNFL circle scans 5 , Liu et al found at least one artefact in 46.3% of the 2,313 scans analysed in their study 6 .…”

Section: Discussionmentioning

confidence: 99%

Quantification of Retinal Nerve Fibre Layer Thickness on Optical Coherence Tomography with a Deep Learning Segmentation-Free Approach

Mariottoni

Jammal

Urata

et al. 2020

Sci Rep

Self Cite

View full text Add to dashboard Cite

This study describes a segmentation-free deep learning (DL) algorithm for measuring retinal nerve fibre layer (RNFL) thickness on spectral-domain optical coherence tomography (SDOCT). The study included 25,285 B-scans from 1,338 eyes of 706 subjects. Training was done to predict RNFL thickness from raw unsegmented scans using conventional RNFL thickness measurements from good quality images as targets, forcing the DL algorithm to learn its own representation of RNFL. The algorithm was tested in three different sets: (1) images without segmentation errors or artefacts, (2) low-quality images with segmentation errors, and (3) images with other artefacts. In test set 1, segmentation-free RNFL predictions were highly correlated with conventional RNFL thickness (r = 0.983, P < 0.001). In test set 2, segmentation-free predictions had higher correlation with the best available estimate (tests with good quality taken in the same date) compared to those from the conventional algorithm (r = 0.972 vs. r = 0.829, respectively; P < 0.001). Segmentation-free predictions were also better in test set 3 (r = 0.940 vs. r = 0.640, P < 0.001). In conclusion, a novel segmentation-free algorithm to extract RNFL thickness performed similarly to the conventional method in good quality images and better in images with errors or other artefacts.

show abstract

Detecting Retinal Nerve Fibre Layer Segmentation Errors on Spectral Domain-Optical Coherence Tomography with a Deep Learning Algorithm

Cited by 22 publications

References 31 publications

Ensemble Deep Learning for Diabetic Retinopathy Detection Using Optical Coherence Tomography Angiography

Ensemble Deep Learning for Diabetic Retinopathy Detection Using Optical Coherence Tomography Angiography

Artificial Intelligence Mapping of Structure to Function in Glaucoma

Quantification of Retinal Nerve Fibre Layer Thickness on Optical Coherence Tomography with a Deep Learning Segmentation-Free Approach

Contact Info

Product

Resources

About