A Line to Align: Deep Dynamic Time Warping for Retinal OCT Segmentation

Maier, Heiko; Faghihroohi, Shahrooz; Navab, Nassir

doi:10.1007/978-3-030-87193-2_67

Cited by 5 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We follow the same experimental protocol for training and evaluation of our method on the Duke OCT dataset, as in prior works [29,18]. The Duke OCT dataset consists of OCT scans from 10 patients, which are annotated by two experts.…”

Section: Methodsmentioning

confidence: 99%

“…Feature Pyramid Networks (FPNs), which are commonly used in the computer vision community, have also been of interest in medical image segmentation for global feature extraction [6,16]. Other lines of work focus on networks designed specifically for the OCT segmentation task [23,31,12], using Gaussian process [22], feature alignment [4,18], or epistemic uncertainty [21].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

$Υ$-Net: A Spatiospectral Network for Retinal OCT Segmentation

Farshad¹,

Yeganeh²,

Gehlbach³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Automated segmentation of retinal optical coherence tomography (OCT) images has become an important recent direction in machine learning for medical applications. We hypothesize that the anatomic structure of layers and their high-frequency variation in OCT images make retinal OCT a fitting choice for extracting spectral domain features and combining them with spatial domain features. In this work, we present Υ-Net, an architecture that combines the frequency domain features with the image domain to improve the segmentation performance of OCT images. The results of this work demonstrate that the introduction of two branches, one for spectral and one for spatial domain features, brings very significant improvement in fluid segmentation performance and allows outperformance as compared to the well-known U-Net model. Our improvement was 13% on the fluid segmentation dice score and 1.9% on the average dice score. Finally, removing selected frequency ranges in the spectral domain demonstrates the impact of these features on the fluid segmentation outperformance.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

$Υ$-Net: A Spatiospectral Network for Retinal OCT Segmentation

Farshad¹,

Yeganeh²,

Gehlbach³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…We evaluated the performance of TransU2Net on the Duke OCT dataset and compared its results with those of other state-of-the-art (SOTA) models, including (A) RelayNet 32 (B) Language 33 (C) Alignment 34 (D)U-Net (E) Y-Net, with all comparative SOTA model results derived from prior work 31,33,34 . It was observed that TransU2Net surpasses the other SOTA models in overall performance, achieving an average Dice of 86.34%, particularly excelling in the segmentation of the NFL-IPL, INL, OPL, and ONL-ISM layers.…”

Section: Comparison With Other Models On the Duke Oct Datasetmentioning

confidence: 99%

TransU2Net: A Hybrid Network for Retinal Layer Segmentation in OCT

Zhu,

Chen,

Shi

et al. 2024

Preprint

View full text Add to dashboard Cite

The retina, a transparent membrane at the eye’s back, is crucial for capturing light signals. OCT provides a non-invasive method for high-resolution imaging of these layers. Its complex, uneven structure, along with OCT imaging low contrast and noise, complicates segmentation for clinical or research applications. Different eye diseases variably affect retinal layers. This paper introduces TransU2Net, a hybrid network for retinal layer segmentation in OCT images. TransU2Net combines U2-Net and Transformer architectures, utilizing U2Net_Half Encoder for feature extraction and a Transformer module to capture global and local features. Additionally, the model incorporates a Convolutional Block Attention Module (CBAM) in its decoder to enhance its performance. We conducted ablation experiments on our collected dataset to demonstrate the respective roles of U2Net_Half_Encoder and CBAM in the encoding and decoding processes. Subsequently, we compared the TransU2Net with other state-of-the-art (SOTA) models. On the private dataset, the average Dice and IoU achieved 91.64% and 84.80%. On the Duke OCT dataset, the average Dice achieved 84.80%, surpassing other SOTA models in both datasets.

show abstract

“…[28] studied the ambiguity in ground truth labels of OCT images and its impact on deep learning-based RLS systems. Among the most recent works, [29] proposed a multi-scale and dual attention-enhanced nested U-Net architecture and [30] modified the dynamic time warping algorithm for an end-to-end layer segmentation with correct topological order. The works by He et al [31,32,33], similar to our proposed method, utilized residual U-Net [34] and performed column-wise soft argmax to get smooth contours.…”

Section: Related Workmentioning

confidence: 99%

Cu-Net: Towards Continuous Multi-Class Contour Detection for Retinal Layer Segmentation In Oct Images

Bhattarai

Kambhamettu

Jin³

2022

2022 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Recent deep learning-based contour detection studies show high accuracy in single-class boundary detection problems. However, this performance does not translate well in a multi-class scenario where continuous contours are required. Our research presents CU-Net, a U-Net-based network with residual-net encoders which can produce accurate and uninterrupted contour lines for multiple classes. The critical factor behind this concept is our continuity module, containing an interpolation layer and a novel activation function that converts discrete signals into smooth contours. We find the application of our approach in medical imaging problems like retinal layer segmentation from optical coherence tomography (OCT) scans. We applied our method to an expert annotated OCT dataset of children with sickle-cell disease. To compare with benchmarks, we evaluated our network on DME and HC-MS datasets. We achieved an overall mean absolute distance of 6.48 ± 2.04µM and 1.97 ± 0.89µM , respectively 1.03 and 1.4 times less than the current state-of-the-art.

show abstract

A Line to Align: Deep Dynamic Time Warping for Retinal OCT Segmentation

Cited by 5 publications

References 20 publications

$Υ$-Net: A Spatiospectral Network for Retinal OCT Segmentation

$Υ$-Net: A Spatiospectral Network for Retinal OCT Segmentation

TransU2Net: A Hybrid Network for Retinal Layer Segmentation in OCT

Cu-Net: Towards Continuous Multi-Class Contour Detection for Retinal Layer Segmentation In Oct Images

Contact Info

Product

Resources

About