Unsupervised Microvascular Image Segmentation Using an Active Contours Mimicking Neural Network

Gur, Shir; Wolf, Lior; Golgher, Lior; Blinder, Pablo

doi:10.1109/iccv.2019.01082

Cited by 48 publications

(38 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…8(a), which presents the correlation coefficients matrix between different axial slices, after removing the diagonal. The situation is not much better when multiplying the raw data by the segmentation mask of [11], since the sparsity of the imaging modality leads to very noisy measurements ( Fig. 6(b) and Fig.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Microvascular Dynamics from 4D Microscopy Using Temporal Segmentation

et al. 2019

Self Cite

View full text Add to dashboard Cite

Recently developed methods for rapid continuous volumetric two-photon microscopy facilitate the observation of neuronal activity in hundreds of individual neurons and changes in blood flow in adjacent blood vessels across a large volume of living brain at unprecedented spatio-temporal resolution. However, the high imaging rate necessitates fully automated image analysis, whereas tissue turbidity and photo-toxicity limitations lead to extremely sparse and noisy imagery. In this work, we extend a recently proposed deep learning volumetric blood vessel segmentation network, such that it supports temporal analysis. With this technology, we are able to track changes in cerebral blood volume over time and identify spontaneous arterial dilations that propagate towards the pial surface. This new capability is a promising step towards characterizing the hemodynamic response function upon which functional magnetic resonance imaging (fMRI) is based.

show abstract

Section: Resultsmentioning

confidence: 99%

“…Our method extends the time-collapsed segmentation [11], where we add an additional novel skeletonization layer on top of the network and perform per-frame segmentation of the 4D movie. The resulting skeleton serves as an anchoring structure that ties all temporal results together regardless of the transient vascular changes.…”

Section: The Acwe Networkmentioning

confidence: 99%

Microvascular Dynamics from 4D Microscopy Using Temporal Segmentation

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…This new loss function combines geometrical information with region similarity hence it provides more precise segmentation. The ACM loss function is used as a loss function in many deep learning models as in [42][43][44][45] .…”

Section: Deep Learning Approachesmentioning

confidence: 99%

COVID-19 Lesion Segmentation using Lung CT Scan Images: Comparative Study based on Active Contour Models

Akbari

Hassen

Al‐Maadeed

et al. 2020

Preprint

View full text Add to dashboard Cite

Abstract Pneumonia is a lung infection threaten that threats all age groups. In this paper, using CT scans images, we used active contour models to evaluate and determine pneumonia infection caused by the Coronavirus disease (COVID-19). A background of active contour models (ACM) including contour representation and object boundary description methods is presented. The focus of this paper is on the conducted works based on the external forces. These methods include edge-based and region-based methods. Furthermore, the explanations of these methods, as well as the advantages and disadvantages of each method are presented. Finally, a comparison between the performances of the conducted works has been done based on a database of Lung CT Scan Images. The present review helps readers identify research starting points in active contour models on COVID19 research, which is a high priority topic to guide researchers and practitioners. In addition, when there are not enough images to use machine learning techniques, such as deep learning methods, the experimental results indicate that active contour methods obtain promising results.

show abstract

“…In recent years, techniques based on deep learning have shown significant improvement over traditional methods for 2PM vascular segmentation [8][9][10]17]. One of the first works in this line was done by Teikari et.…”

Section: Introductionmentioning

confidence: 99%

“…al. [9] recently proposed an unsupervised DNN based on the active contours method, and demonstrated improved generalization capability compared to supervised models [10,17,18], with faster segmentation speed. However, this method still suffers from excessive training and inference times, and high computational cost.…”

Section: Introductionmentioning

confidence: 99%

Anatomical modeling of brain vasculature in two-photon microscopy by generalizable deep learning

Tahir

Kura

Zhu

et al. 2020

Preprint

View full text Add to dashboard Cite

Segmentation of blood vessels from two-photon microscopy (2PM) angiograms of brains has important applications in hemodynamic analysis and disease diagnosis. Here we develop a generalizable deep-learning technique for accurate 2PM vascular segmentation of sizable regions in mouse brains acquired from multiple 2PM setups. In addition, the technique is computationally efficient, making it ideal for large-scale neurovascular analysis. Introduction: Vascular segmentation from 2PM angiograms is usually an important first step in hemodynamic modeling of brain vasculature. Existing state-of-the-art segmentation methods based on deep learning either lack the ability to generalize to data from various imaging systems, or are computationally infeasible for large-scale angiograms. In this work, we present a method which improves upon both these limitations by being generalizable to various imaging systems, and also being able to segment very large-scale angiograms. Methods: We employ a computationally efficient deep learning framework based on a semi-supervised learning strategy, whose effectiveness we demonstrate on experimentally acquired in-vivo angiograms from mouse brains of dimensions up to 808x808x702 micrometers. Results: After training on data from only one 2PM microscope, we perform vascular segmentation on data from another microscope without any network tuning. Our method demonstrates 10x faster computation in terms of voxels-segmented-persecond and 3x larger depth compared to the state-of-the-art. Conclusion: Our work provides a generalizable and computationally efficient anatomical modeling framework for the brain vasculature, which consists of deeplearning based vascular segmentation followed by graphing. It paves the way for future modeling and analysis of hemodynamic response at much greater scales that were inaccessible before.

show abstract

Unsupervised Microvascular Image Segmentation Using an Active Contours Mimicking Neural Network

Cited by 48 publications

References 39 publications

Microvascular Dynamics from 4D Microscopy Using Temporal Segmentation

Microvascular Dynamics from 4D Microscopy Using Temporal Segmentation

COVID-19 Lesion Segmentation using Lung CT Scan Images: Comparative Study based on Active Contour Models

Anatomical modeling of brain vasculature in two-photon microscopy by generalizable deep learning

Contact Info

Product

Resources

About