Fusing Results of Several Deep Learning Architectures for Automatic Classification of Normal and Diabetic Macular Edema in Optical Coherence Tomography

Chan, Genevieve C. Y.; Kamble, Ravi; Müller, Henning; Shah, Syed Ayaz Ali; Tang, Tong Boon; Mériaudeau, Fabrice

doi:10.1109/embc.2018.8512371

Cited by 21 publications

(11 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CME is caused by cystoid change due to liquefactive necrosis of Müller cells on the basis of DRT. In addition to VEGF, its pathogenesis is also related to prostaglandins and in ammatory cytokines [7]. The pathogenesis of the SRD type remains unclear.…”

Section: Discussionmentioning

confidence: 99%

Efficacy of Aflibercept for DME with Different OCT Types

Tao

Kou

Xia³

et al. 2021

Preprint

View full text Add to dashboard Cite

Background: This study aimed to assess efficacy differences of aflibercept by intravitreal injection for diabetic macular edema (DME) with different OCT types. Methods: according to OCT morphology, 128 patients (128 eyes) with DME were divided into the diffuse retinal thickening (DRT; 41 patients, 41 eyes), cystoid macular edema (CME; 47 patients, 47 eyes), and serous retinal detachment (SRD; 40 patients, 40eyes) groups. All patients were treated by intravitreal injection of 0.5 mg (0.05 mL) aflibercept according to the 3+prn regimen. BCVA (LogMAR) and central foveal thickness (CFT) among the three groups at baseline, and 1 month, 3 months, 6 months and 12 months after treatment, respectively, were assessed. Results: At 1 month, 3 months, 6 months and 12 months after treatment, improvements in BCVA and CFT in the three groups showed statistical significance (P<0.05). BCVA improvement was better at all time points in the DRT group compared with the SRD group (P<0.05). BCVA improvement was also better in the CME group compared with the SRD group at 6 months and 12 months post-treatment (P<0.05). At 1 month, 3 months and 6 months after treatment, CFT differences among the three groups had no statistical significance (P>0.05), while at 12 months post-treatment, CFT values among the three groups had significant differences (F=3.77, P=0.0269). at 12 months after treatment, CFT improvement was better in the DRT and CME groups compared with the SRD group (P<0.05). Conclusion: aflibercept by intravitreal injection could significantly improve visual acuity in DME patients with different OCT types and reduce CFT. At the initial treatment of 12 months, CFT and BCVA improvements in the DRT and CME types were the most pronounced.

show abstract

Section: Discussionmentioning

confidence: 99%

Efficacy of Aflibercept for DME with Different OCT Types

Tao

Kou

Xia³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…The primary goal of the study was to distinguish between the DME case and the normal case. 16 Two frameworks for detecting retinal disorders were presented. The first was created by combining features extracted from the pretrained CNN architectures, such as AlexNet, VGG Net, and GoogleNet.…”

Section: Related Workmentioning

confidence: 99%

“…The VGG‐16 achieved a precision of 93.01%. The primary goal of the study was to distinguish between the DME case and the normal case 16 …”

Section: Related Workmentioning

confidence: 99%

Transfer learning‐based platform for detecting multi‐classification retinal disorders using optical coherence tomography images

Saleh

Wahed

Salaheldin

2021

Int J Imaging Syst Tech

View full text Add to dashboard Cite

One of the primary concerns of computer‐aided diagnosis is the detection of retinal disorders. The study aims to categorize the patients into choroidal neovascularization, diabetic macular edema, drusen, and normal by using optical coherence tomography (OCT) images. For the first time, two novel transfer learning‐based techniques were used for retinal disorder classification: SqueezeNet and the Inception V3 Net. Two SqueezeNet scenarios were used to compare the performance of the original SqueezeNet and the improved one. A dataset of 11 200 OCT images was used for data partitioning of SqueezeNet and, meanwhile, 18 000 images for Inception V3 Net. The modified SqueezeNet achieved 98% accuracy, a 1.2% improvement over the original. The Inception V3 Net classifier improved its classification accuracy to 98.4%. When compared to other classifiers and a human expert, the transfer learning approach demonstrated its robustness in the challenge of retinal disorders classification with a large dataset.

show abstract

“…Chan et al [44] presented a framework on a dataset of fundus images obtained from various hospitals by incorporating both domain knowledge and features learned from a deep learning model. This method was also used in other applications [45].…”

Section: Application and Evaluation On Earlier Glaucoma Screening And Diagnosis Classificationmentioning

confidence: 99%

Detection of retinal abnormalities in fundus image using CNN deep learning networks

Akil

Elloumi

Kachouri

2021

State of the Art in Neural Networks and Their Applications

View full text Add to dashboard Cite

The World Health Organization (WHO) estimates that 285 million people are visually impaired worldwide, with 39 million blinds. Glaucoma, Cataract, Age-related macular degeneration, Diabetic retinopathy are among the leading retinal diseases. Thus, there is an active effort to create and develop methods to automate screening of retinal diseases. Many Computer Aided Diagnosis (CAD) systems for ocular diseases have been developed and are widely used. Deep learning (DL) has shown its capabilities in field of public health including ophthalmology. In retinal disease diagnosis, the approach based upon DL and convolutional neural networks (CNNs) is used to locate, identify, quantify pathological features. The performance of this approach keeps growing. This chapter, addresses an overview of the used methods based upon DL and CNNs in detection of retinal abnormalities related to the most severe ocular diseases in retinal images, where network architectures, post/preprocessing and evaluation experiments are detailed. We also present some related work concerning the Deep Learning-based Smartphone applications for earlier screening and diagnosisof retinal diseases.

show abstract

Fusing Results of Several Deep Learning Architectures for Automatic Classification of Normal and Diabetic Macular Edema in Optical Coherence Tomography

Cited by 21 publications

References 11 publications

Efficacy of Aflibercept for DME with Different OCT Types

Efficacy of Aflibercept for DME with Different OCT Types

Transfer learning‐based platform for detecting multi‐classification retinal disorders using optical coherence tomography images

Detection of retinal abnormalities in fundus image using CNN deep learning networks

Contact Info

Product

Resources

About