RetFluidNet: Retinal Fluid Segmentation for SD-OCT Images Using Convolutional Neural Network

Sappa, Loza Bekalo; Okuwobi, Idowu Paul; Li, Mingchao; Zhang, Yuhan; Xie, Sha; Yuan, Songtao; Tao, Yu-Hui

doi:10.1007/s10278-021-00459-w

Cited by 29 publications

(13 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Regarding the technical implementation, we considered each of our single components to be of high precision: our network for SRF segmentation in OCT scans reaches a Dice score of 0.853. This score is similar to comparable state-of-the-art architectures predicting SRF for CSCR (0.910 [ 49 ]) or more general for macular edema (0.845 [ 50 ], 0.75 [ 51 ], 0.958 [ 52 ]). Our projected en face Dice score as relevant for our usecase is still higher with 0.897.…”

Section: Technical Discussionsupporting

confidence: 67%

Chronological Registration of OCT and Autofluorescence Findings in CSCR: Two Distinct Patterns in Disease Course

et al. 2022

View full text Add to dashboard Cite

Optical coherence tomography (OCT) and fundus autofluorescence (FAF) are important imaging modalities for the assessment and prognosis of central serous chorioretinopathy (CSCR). However, setting the findings from both into spatial and temporal contexts as desirable for disease analysis remains a challenge due to both modalities being captured in different perspectives: sparse three-dimensional (3D) cross sections for OCT and two-dimensional (2D) en face images for FAF. To bridge this gap, we propose a visualisation pipeline capable of projecting OCT labels to en face image modalities such as FAF. By mapping OCT B-scans onto the accompanying en face infrared (IR) image and then registering the IR image onto the FAF image by a neural network, we can directly compare OCT labels to other labels in the en face plane. We also present a U-Net inspired segmentation model to predict segmentations in unlabeled OCTs. Evaluations show that both our networks achieve high precision (0.853 Dice score and 0.913 Area under Curve). Furthermore, medical analysis performed on exemplary, chronologically arranged CSCR progressions of 12 patients visualized with our pipeline indicates that, on CSCR, two patterns emerge: subretinal fluid (SRF) in OCT preceding hyperfluorescence (HF) in FAF and vice versa.

show abstract

Section: Technical Discussionsupporting

confidence: 67%

Chronological Registration of OCT and Autofluorescence Findings in CSCR: Two Distinct Patterns in Disease Course

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Contrary to the order, however, we can observe a dependency solely due to the characteristics of the dataset. The related works that used the publicly available RETOUCH 54 test dataset all report relatively small differences in Dice score between the classes with a std of 0.020 to 0.025 55 – 58 while the ones using private datasets report larger values of 0.075 59 and 0.087 60 . Consequently, we assign the reason for the spread of the Dice scores between the classes primarily to the characteristic of the dataset.…”

Section: Discussionmentioning

confidence: 98%

“…However, with the design of increasingly advanced architectures 14 – 16 , recent single-stage networks can learn the information induced through pre- or post-processing by themselves. This comes with the advantage of reduced computational complexity and, consequently, making these methods more clinically applicable 17 . Despite the advances provided by DL, there is still a need for more accurate automatic segmentation tools.…”

Section: Introductionmentioning

confidence: 99%

A vision transformer architecture for the automated segmentation of retinal lesions in spectral domain optical coherence tomography images

Philippi

Rothaus

Castelli

2023

Sci Rep

View full text Add to dashboard Cite

Neovascular age-related macular degeneration (nAMD) is one of the major causes of irreversible blindness and is characterized by accumulations of different lesions inside the retina. AMD biomarkers enable experts to grade the AMD and could be used for therapy prognosis and individualized treatment decisions. In particular, intra-retinal fluid (IRF), sub-retinal fluid (SRF), and pigment epithelium detachment (PED) are prominent biomarkers for grading neovascular AMD. Spectral-domain optical coherence tomography (SD-OCT) revolutionized nAMD early diagnosis by providing cross-sectional images of the retina. Automatic segmentation and quantification of IRF, SRF, and PED in SD-OCT images can be extremely useful for clinical decision-making. Despite the excellent performance of convolutional neural network (CNN)-based methods, the task still presents some challenges due to relevant variations in the location, size, shape, and texture of the lesions. This work adopts a transformer-based method to automatically segment retinal lesion from SD-OCT images and qualitatively and quantitatively evaluate its performance against CNN-based methods. The method combines the efficient long-range feature extraction and aggregation capabilities of Vision Transformers with data-efficient training of CNNs. The proposed method was tested on a private dataset containing 3842 2-dimensional SD-OCT retina images, manually labeled by experts of the Franziskus Eye-Center, Muenster. While one of the competitors presents a better performance in terms of Dice score, the proposed method is significantly less computationally expensive. Thus, future research will focus on the proposed network’s architecture to increase its segmentation performance while maintaining its computational efficiency.

show abstract

“…Contrary to the order, however, we can observe a dependency solely due to the characteristics of the dataset. The related works that used the publicly available RETOUCH 47 test dataset all report relatively small differences in Dice score between the classes with a std of 0.020 to 0.025 [48][49][50][51] while the ones using private datasets report larger values of 0.075 52 and 0.087 53 . Consequently, we assign the reason for the spread of the Dice scores between the classes primarily to the characteristic of the dataset.…”

Section: Class-wise Performancementioning

confidence: 98%

“…However, with the design of increasingly advanced architectures [12][13][14] , recent single-stage networks can learn the information induced through pre-or post-processing by themselves. This comes with the advantage of reduced computational complexity and, consequently, making these methods more clinically applicable 15 . Despite the advances provided by DL, there is still a need for more accurate automatic segmentation tools.…”

mentioning

confidence: 99%

A Vision Transformer Architecture for the Automated Segmentation of Retinal Fluids in Spectral Domain Optical Coherence Tomography Images

Philippi

Rothaus

Castelli

2022

Preprint

View full text Add to dashboard Cite

Neovascular age-related macular degeneration (nAMD) is one of the major causes of irreversible blindness and is characterized by accumulations of different fluids inside the retina. An early detection and activity monitoring of predominately three types of fluids, namely intra-retinal fluid (IRF), sub-retinal fluid (SRF), and pigment epithelium detachment (PED), is critical for a successful treatment. Spectral-domain optical coherence tomography (SD-OCT) revolutionized nAMD treatment by providing cross-sectional, high-resolution images of the retina. Automatic segmentation and quantification of IRF, SRF, and PED in SD-OCT images can be extremely useful for clinical decision-making. Despite the use of state-of-the-art convolutional neural network (CNN)-based methods, the task remains challenging due to relevant variations in the location, size, shape, and texture of the fluids. This work is the first to adopt a transformer-based method to automatically segment retinal fluid from SD-OCT images and qualitatively and quantitatively evaluate its performance against CNN-based methods. The method combines the efficient long-range feature extraction and aggregation capabilities of Vision Transformers (ViTs) with data-efficient training of CNNs. The proposed method was tested on a private dataset containing 3842 2-dimensional SD-OCT retina images, manually labeled by experts of the Franziskus-Eye-Hospital. While one of the competitors presents a better performance in terms of Dice score, the proposed method is significantly less computationally expensive. Thus, future research will focus on the proposed network's architecture to increase its segmentation performance while maintaining its computational efficiency.

show abstract

RetFluidNet: Retinal Fluid Segmentation for SD-OCT Images Using Convolutional Neural Network

Cited by 29 publications

References 37 publications

Chronological Registration of OCT and Autofluorescence Findings in CSCR: Two Distinct Patterns in Disease Course

Chronological Registration of OCT and Autofluorescence Findings in CSCR: Two Distinct Patterns in Disease Course

A vision transformer architecture for the automated segmentation of retinal lesions in spectral domain optical coherence tomography images

A Vision Transformer Architecture for the Automated Segmentation of Retinal Fluids in Spectral Domain Optical Coherence Tomography Images

Contact Info

Product

Resources

About