An Improved MPB-CNN Segmentation Method for Edema Area and Neurosensory Retinal Detachment in SD-OCT Images

Jian, Fang Fang; Zhang, Yuhan; Xie, Keren; Yuan, Songtao; Chen, Qiang

doi:10.1007/978-3-030-32956-3_16

Cited by 6 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The performance of RetFluidNet was evaluated against existing methods named Deeplabv3 [41], fully convolutional network (FCN) [42], UNet++ [41], and an improved multi-scale parallel branch convolutional neural network (Im-MPB-CNN) [35]. RetFluidNet and comparative methods were set to have similar learning rate values, batch size, and the number of training iterations.…”

Section: Comparative Methods and Evaluation Metricsmentioning

confidence: 99%

“…Remember that we have integrated these concepts in different architecture to produce plausibly good results. Comparison against Im-MPB-CNN [35] is included to evaluate RetFluidNet with the method specifically designed for AMD fluid segmentation.…”

Section: Comparative Methods and Evaluation Metricsmentioning

confidence: 99%

“…The convolutional neural network (CNN) introduction enabled advancing retinal fluid segmentation [21,[21][22][23][24][25][26][27][28][29][30][31][32][33][34][35][36]. A fully convolutional neural network (FCN)-based model presented in Roy et al [21] uses the image data and distance map to segment retinal fluids.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

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

Sappa

Okuwobi

et al. 2021

J Digit Imaging

Self Cite

View full text Add to dashboard Cite

Age-related macular degeneration (AMD) is one of the leading causes of irreversible blindness and is characterized by fluidrelated accumulations such as intra-retinal fluid (IRF), subretinal fluid (SRF), and pigment epithelial detachment (PED). Spectral-domain optical coherence tomography (SD-OCT) is the primary modality used to diagnose AMD, yet it does not have algorithms that directly detect and quantify the fluid. This work presents an improved convolutional neural network (CNN)-based architecture called RetFluidNet to segment three types of fluid abnormalities from SD-OCT images. The model assimilates different skip-connect operations and atrous spatial pyramid pooling (ASPP) to integrate multi-scale contextual information; thus, achieving the best performance. This work also investigates between consequential and comparatively inconsequential hyperparameters and skip-connect techniques for fluid segmentation from the SD-OCT image to indicate the starting choice for future related researches. RetFluidNet was trained and tested on SD-OCT images from 124 patients and achieved an accuracy of 80.05%, 92.74%, and 95.53% for IRF, PED, and SRF, respectively. RetFluidNet showed significant improvement over competitive works to be clinically applicable in reasonable accuracy and time efficiency. RetFluidNet is a fully automated method that can support early detection and follow-up of AMD.

show abstract

Section: Comparative Methods and Evaluation Metricsmentioning

confidence: 99%

Section: Comparative Methods and Evaluation Metricsmentioning

confidence: 99%

See 1 more Smart Citation

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

Sappa

Okuwobi

et al. 2021

J Digit Imaging

Self Cite

View full text Add to dashboard Cite

show abstract

“…We first presented UNet++ in our DLMIA 2018 paper [51]. UNet++ has since been quickly adopted by the research community, either as a strong baseline for comparison [52], [53], [54], [55], or as a source of inspiration for developing newer semantic segmentation architectures [56], [57], [58], [59], [60], [61]; it has also been utilized for multiple applications, such as segmenting objects in biomedical images [62], [63], natural images [64], and satellite images [65], [66]. Recently, Shenoy [67] has independently and systematically investigated UNet++ for the task of "contact prediction model PconsC4", demonstrating significant improvement over widely-used U-Net.…”

Section: Our Previous Workmentioning

confidence: 99%

UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation

Zhou

Siddiquee

Tajbakhsh

et al. 2020

IEEE Trans. Med. Imaging

2,606

1,038

View full text Add to dashboard Cite

The state-of-the-art models for medical image segmentation are variants of U-Net and fully convolutional networks (FCN). Despite their success, these models have two limitations:(1) their optimal depth is apriori unknown, requiring extensive architecture search or inefficient ensemble of models of varying depths; and (2) their skip connections impose an unnecessarily restrictive fusion scheme, forcing aggregation only at the samescale feature maps of the encoder and decoder sub-networks. To overcome these two limitations, we propose UNet++, a new neural architecture for semantic and instance segmentation, by (1) alleviating the unknown network depth with an efficient ensemble of U-Nets of varying depths, which partially share an encoder and co-learn simultaneously using deep supervision; (2) redesigning skip connections to aggregate features of varying semantic scales at the decoder sub-networks, leading to a highly flexible feature fusion scheme; and (3) devising a pruning scheme to accelerate the inference speed of UNet++. We have evaluated UNet++ using six different medical image segmentation datasets, covering multiple imaging modalities such as computed tomography (CT), magnetic resonance imaging (MRI), and electron microscopy (EM), and demonstrating that (1) UNet++ consistently outperforms the baseline models for the task of semantic segmentation across different datasets and backbone architectures; (2) UNet++ enhances segmentation quality of varying-size objects-an improvement over the fixed-depth U-Net; (3) Mask RCNN++ (Mask R-CNN with UNet++ design) outperforms the original Mask R-CNN for the task of instance segmentation; and (4) pruned UNet++ models achieve significant speedup while showing only modest performance degradation. Our implementation and pre-trained models are available at https://github.com/MrGiovanni/UNetPlusPlus.

show abstract

“…In ophthalmology, there is also an emerging study of applying AI to analyze ultrasound images [30], retinal fundus images [31], and OCT [11,15,19,[32][33][34][35][36]. More recently, there are emerging reports of utilizing AI algorithms in OCT images to diseases such as diabetic retinopathy [37], glaucoma [11,38], macular degeneration [36], and retinal detachment [39]. However, there is still a lack of AI studies for the diagnosis of ERM.…”

Section: Introductionmentioning

confidence: 99%

Automated Detection of Epiretinal Membranes in OCT Images Using Deep Learning

et al. 2022

View full text Add to dashboard Cite

Introduction: Development and validation of a deep learning algorithm to automatedly identify and locate ERM regions in OCT images. Methods: OCT images of 468 eyes were retrospectively collected from a total of 404 ERM patients. One expert manually annotated the ERM regions for all images. A total of 422 images (90%) and the rest 46 images (10%) were used as the training dataset and validation dataset for deep learning algorithm training and validation, respectively. One senior and one junior clinician read the images. The diagnostic results were compared. Results: The algorithm accurately segmented and located the ERM regions in OCT images. The image-level accuracy was 95.65%, and the ERM region-level accuracy was 90.14%, respectively. In comparison experiments, the accuracies of the junior clinician improved from 85.00% and 61.29% without the assistance of the algorithm to 100.00% and 90.32% with the assistance of the algorithm. The corresponding results of the senior clinician were 96.15%, 95.00% without the assistance of the algorithm, and 96.15%, 97.50% with the assistance of the algorithm. Conclusions: The developed deep learning algorithm can accurately segmenting ERM regions in OCT images. This deep learning approach may help clinicians in clinical diagnosis with better accuracy and efficiency.

show abstract

An Improved MPB-CNN Segmentation Method for Edema Area and Neurosensory Retinal Detachment in SD-OCT Images

Cited by 6 publications

References 10 publications

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

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

UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation

Automated Detection of Epiretinal Membranes in OCT Images Using Deep Learning

Contact Info

Product

Resources

About