Tags

Type your tag names separated by a space and hit enter

Ensemble Deep Learning for Diabetic Retinopathy Detection Using Optical Coherence Tomography Angiography.
Transl Vis Sci Technol. 2020 04; 9(2):20.TV

Abstract

Purpose

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 fine-tuned 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.

Authors+Show Affiliations

School of Engineering Science, Simon Fraser University, Burnaby, British Columbia, Canada.Department of Ophthalmology and Visual Sciences, University of British Columbia, Vancouver, British Columbia, Canada.School of Engineering Science, Simon Fraser University, Burnaby, British Columbia, Canada.Department of Ophthalmology and Visual Sciences, University of British Columbia, Vancouver, British Columbia, Canada.School of Engineering Science, Simon Fraser University, Burnaby, British Columbia, Canada.Department of Ophthalmology and Visual Sciences, University of British Columbia, Vancouver, British Columbia, Canada.Department of Ophthalmology and Visual Sciences, University of British Columbia, Vancouver, British Columbia, Canada.School of Engineering Science, Simon Fraser University, Burnaby, British Columbia, Canada.Department of Ophthalmology and Visual Sciences, University of British Columbia, Vancouver, British Columbia, Canada.School of Engineering Science, Simon Fraser University, Burnaby, British Columbia, Canada.

Pub Type(s)

Journal Article

Language

eng

PubMed ID

32818081

Citation

Heisler, Morgan, et al. "Ensemble Deep Learning for Diabetic Retinopathy Detection Using Optical Coherence Tomography Angiography." Translational Vision Science & Technology, vol. 9, no. 2, 2020, p. 20.
Heisler M, Karst S, Lo J, et al. Ensemble Deep Learning for Diabetic Retinopathy Detection Using Optical Coherence Tomography Angiography. Transl Vis Sci Technol. 2020;9(2):20.
Heisler, M., Karst, S., Lo, J., Mammo, Z., Yu, T., Warner, S., Maberley, D., Beg, M. F., Navajas, E. V., & Sarunic, M. V. (2020). Ensemble Deep Learning for Diabetic Retinopathy Detection Using Optical Coherence Tomography Angiography. Translational Vision Science & Technology, 9(2), 20. https://doi.org/10.1167/tvst.9.2.20
Heisler M, et al. Ensemble Deep Learning for Diabetic Retinopathy Detection Using Optical Coherence Tomography Angiography. Transl Vis Sci Technol. 2020;9(2):20. PubMed PMID: 32818081.
* Article titles in AMA citation format should be in sentence-case
TY - JOUR T1 - Ensemble Deep Learning for Diabetic Retinopathy Detection Using Optical Coherence Tomography Angiography. AU - Heisler,Morgan, AU - Karst,Sonja, AU - Lo,Julian, AU - Mammo,Zaid, AU - Yu,Timothy, AU - Warner,Simon, AU - Maberley,David, AU - Beg,Mirza Faisal, AU - Navajas,Eduardo V, AU - Sarunic,Marinko V, Y1 - 2020/04/13/ PY - 2019/10/01/received PY - 2020/01/23/accepted PY - 2020/8/21/entrez PY - 2020/8/21/pubmed PY - 2020/8/21/medline KW - deep learning KW - diabetic retinopathy KW - machine learning KW - optical coherence tomography KW - optical coherence tomography angiography SP - 20 EP - 20 JF - Translational vision science & technology JO - Transl Vis Sci Technol VL - 9 IS - 2 N2 - Purpose: 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 fine-tuned 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. SN - 2164-2591 UR - https://www.unboundmedicine.com/medline/citation/32818081/Ensemble_Deep_Learning_for_Diabetic_Retinopathy_Detection_Using_Optical_Coherence_Tomography_Angiography_ L2 - https://tvst.arvojournals.org/article.aspx?doi=10.1167/tvst.9.2.20 DB - PRIME DP - Unbound Medicine ER -