# A deep learning network with differentiable dynamic programming for retina OCT surface segmentation

arXiv: https://arxiv.org/abs/2210.06335 · HTML: https://mlsanity.com/paper/2210.06335/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 31 references that are in the map.
- First posted on arXiv 8 Oct 2022.

## Builds on (most cited first, 9 of 9)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [A Survey on Deep Learning in Medical Image Analysis](https://mlsanity.com/paper/1702.05747/) (2017)
- [Deep Learning for Medical Image Analysis](https://mlsanity.com/paper/1708.08987/) (2017)
- [Differentiable Dynamic Programming for Structured Prediction and Attention](https://mlsanity.com/paper/1802.03676/) (2018)
- [ReLayNet: Retinal Layer and Fluid Segmentation of Macular Optical Coherence Tomography using Fully Convolutional Network](https://mlsanity.com/paper/1704.02161/) (2017)
- [Simultaneous Alignment and Surface Regression Using Hybrid 2D-3D Networks for 3D Coherent Layer Segmentation of Retina OCT Images](https://mlsanity.com/paper/2203.02390/) (2022)
- [Probabilistic Intra-Retinal Layer Segmentation in 3-D OCT Images Using Global Shape Regularization](https://mlsanity.com/paper/1403.8003/) (2014)
- [Optimal Surface Segmentation with Convex Priors in Irregularly Sampled Space](https://mlsanity.com/paper/1611.03059/) (2016)
- [Simultaneous Multiple Surface Segmentation Using Deep Learning](https://mlsanity.com/paper/1705.07142/) (2017)

## Built on by (most cited first, 2 of 2)

- [gcDLSeg: Integrating Graph-cut into Deep Learning for Binary Semantic Segmentation](https://mlsanity.com/paper/2312.04713/) (2023)
- [Formula-Driven Data Augmentation and Partial Retinal Layer Copying for Retinal Layer Segmentation](https://mlsanity.com/paper/2410.01185/) (2024)

## Abstract (arXiv)

Multiple-surface segmentation in Optical Coherence Tomography (OCT) images is a challenge problem, further complicated by the frequent presence of weak image boundaries. Recently, many deep learning (DL) based methods have been developed for this task and yield remarkable performance. Unfortunately, due to the scarcity of training data in medical imaging, it is challenging for DL networks to learn the global structure of the target surfaces, including surface smoothness. To bridge this gap, this study proposes to seamlessly unify a U-Net for feature learning with a constrained differentiable dynamic programming module to achieve an end-to-end learning for retina OCT surface segmentation to explicitly enforce surface smoothness. It effectively utilizes the feedback from the downstream model optimization module to guide feature learning, yielding a better enforcement of global structures of the target surfaces. Experiments on Duke AMD (age-related macular degeneration) and JHU MS (multiple sclerosis) OCT datasets for retinal layer segmentation demonstrated very promising segmentation accuracy.