LIFE: A Generalizable Autodidactic Pipeline for 3D OCT-A Vessel Segmentation

  title={LIFE: A Generalizable Autodidactic Pipeline for 3D OCT-A Vessel Segmentation},
  author={Dewei Hu and Can Cui and Hao Li and Kathleen E. Larson and Yuankai K. Tao and Ipek Oguz},
  journal={Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention},
  • Dewei Hu, C. Cui, I. Oguz
  • Published 9 July 2021
  • Computer Science
  • Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
Optical coherence tomography (OCT) is a non-invasive imaging technique widely used for ophthalmology. It can be extended to OCT angiography (OCT-A), which reveals the retinal vasculature with improved contrast. Recent deep learning algorithms produced promising vascular segmentation results; however, 3D retinal vessel segmentation remains difficult due to the lack of manually annotated training data. We propose a learning-based method that is only supervised by a self-synthesized modality named… 
Domain Generalization for Retinal Vessel Segmentation with Vector Field Transformer
A novel pipeline is investigated that generalizes the retinal vessel segmentation across color fundus photography and OCT angiography images by hypothesizing that the scaled minor eigenvector of the Hessian matrix can sufficiently represent the vessel by vector flow.
Unsupervised denoising of retinal OCT with diffusion probabilistic model
A diffusion probabilistic model that is fully unsupervised to learn from noise instead of signal is presented, which can significantly improve the image quality with a simple working pipeline and a small amount of training data.


Automated and Network Structure Preserving Segmentation of Optical Coherence Tomography Angiograms
This study generates the first open dataset of retinal parafoveal OCTA images with associated ground truth manual segmentations and establishes a standard for OCTA image segmentation by surveying a broad range of state-of-the-art vessel enhancement and binarisation procedures, providing the most comprehensive comparison of these methods under a unified framework to date.
Retinal OCT Denoising with Pseudo-Multimodal Fusion Network
A learning-based method that exploits information from the single-frame noisy B-scan and a pseudo-modality that is created with the aid of the self-fusion method can effectively suppress the speckle noise and enhance the contrast between retina layers while the overall structure and small blood vessels are preserved.
Variational intensity cross channel encoder for unsupervised vessel segmentation on OCT angiography
This work presents a neural network architecture that segments vascular structures in retinal OCTA images without the need of direct supervision and proposes a variational intensity cross channel encoder that finds vessel masks by exploiting the common underlying structure shared by two OCTa images of the the same region but acquired on different devices.
A Deep Learning Approach to Denoise Optical Coherence Tomography Images of the Optic Nerve Head
The deep learning algorithm can denoise a single-frame OCT B-scan of the ONH in under 20 ms, thus offering a framework to obtain superior quality OCT B -scans with reduced scanning times and minimal patient discomfort.
3D Shape Modeling and Analysis of Retinal Microvasculature in OCT-Angiography Images
A robust, effective, and automatic 3D shape modeling framework to provide a high-quality 3D vessel representation and to preserve valuable 3D geometric and topological information for vessel analysis is proposed.
Deep neural ensemble for retinal vessel segmentation in fundus images towards achieving label-free angiography
This paper employs unsupervised hierarchical feature learning using ensemble of two level of sparsely trained denoised stacked autoencoder and shows that ensemble training of auto-encoders fosters diversity in learning dictionary of visual kernels for vessel segmentation.
A texture-based 3D region growing approach for segmentation of ICA through the skull base in CTA
A texture-based 3D region growing approach is proposed and applied to the internal carotid artery segmentation through the skull base that decreases significantly explosions, over-segmentations and increases rate of area overlap, sensitivity, precision at skull base.
Vascular Extraction Using MRA Statistics and Gradient Information
Experimental results on clinical cerebral data demonstrate that using gradient information as a further step improves the mixture model based segmentation of cerebral vasculature, in particular segmentsation of the low contrast vasculatures.
CURVES: Curve evolution for vessel segmentation
Statistical model for OCT image denoising.
A new OCT denoising algorithm is introduced that combines a novel speckle noise model, derived from local statistics of empirical spectral domain OCT (SD-OCT) data, with a Huber variant of total variation regularization for edge preservation.