3D segmentation of abdominal CT imagery with graphical models, conditional random fields and learning

Bhole, Chetan; Pal, Christopher; Rim, David; Wismüller, Axel

doi:10.1007/s00138-013-0497-x

Cited by 23 publications

(5 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Whereas GPU greatly speeds up tasks that involve many relatively small computations, CPU can perform more complex tasks that require larger memory (Hockney and Jesshope (1981); Owens et al (2008); Yu et al (2014)). Another technique is to replace the Gibbs sampler by a faster algorithm for calculating MAP (maximum a posteriori) estimates of the states; see Greig, Porteous and Seheult (1989), Boykov and Kolmogorov (2004), Ravikumar and Lafferty (2006), Kumar and Zilberstein (2011), and Bhole et al (2014). For example, Greig, Porteous and Seheult (1989), Boykov and Kolmogorov (2004) and Bhole et al (2014) reformulate MAP estimation as the solution to a minimum-cut/maximum-flow algorithm on a graph, whose computational complexity in computing the MAP estimates of the states in the H n K n blocks in Section 3 is O ( n 4 / (H n K n ) 3 ) .…”

Section: Discussionmentioning

confidence: 99%

Asymptotically efficient parameter estimation in hidden Markov spatio-temporal random fields

Lai¹,

Lim²

2015

STAT SINICA

View full text Add to dashboard Cite

Estimation of the parameters of Markov random field models for spatial and temporal data arises in many applications. There are computational and statistical challenges in developing efficient estimators because of the complexity of the joint distribution of the spatio-temporal models, especially when they involve hidden states that also need to be estimated from the observations. We develop composite likelihood estimators that are analytically and computationally tractable, and show that they are asymptotically efficient under some mild correlation decay assumptions.

show abstract

Section: Discussionmentioning

confidence: 99%

Asymptotically efficient parameter estimation in hidden Markov spatio-temporal random fields

Lai¹,

Lim²

2015

STAT SINICA

View full text Add to dashboard Cite

show abstract

“…An effective strategy to bring high-level information into graph-based models has been the incorporation of probabilistic atlases (Freedman and Zhang, 2005;Linguraru et al, 2012;Park et al, 2010;Song et al, 2006). Bhole et al (Bhole et al, 2014) used Gaussian mixture models (GMMs) to encode high level semantic information into different graph-based models, including MRFs, Conditional random fields (CRFs) and kernel CRFs.. In (Linguraru et al, 2012), Linguraru et al used abdominal probabilistic atlases and Parzen shape windows (Parzen, 1962) to encode, respectively, spatial and shape constraints directly into the 4D graph-cut-based segmentation framework, as new terms of the energy function.…”

Section: Graph-based Modelsmentioning

confidence: 99%

Computational anatomy for multi-organ analysis in medical imaging: A review

Cerrolaza

Picazo

Humbert³

et al. 2019

Medical Image Analysis

View full text Add to dashboard Cite

The medical image analysis field has traditionally been focused on the development of organ-, and disease-specific methods. Recently, the interest in the development of more comprehensive computational anatomical models has grown, leading to the creation of multi-organ models. Multi-organ approaches, unlike traditional organ-specific strategies, incorporate inter-organ relations into the model, thus leading to a more accurate representation of the complex human anatomy. Inter-organ relations are not only spatial, but also functional and physiological. Over the years, the strategies proposed to efficiently model multi-organ structures have evolved from the simple global modeling, to more sophisticated approaches such as sequential, hierarchical, or machine learning-based models. In this paper, we present a review of the state of the art on multi-organ analysis and associated computation anatomy methodology. The manuscript follows a methodology-based classification of the different techniques available for the analysis of multi-organs and multi-anatomical structures, from techniques using point distribution models to the most recent deep learning-based approaches. With more than 300 papers included in this review, we reflect on the trends and challenges of the field of computational anatomy, the particularities of each anatomical region, and the potential of multi-organ analysis to increase the impact of medical imaging applications on the future of healthcare.

show abstract

“…There has been a large amount of interest in applying CRFs to many different problems. Successful applications have included text processing (Settles, 2005), bioinformatics (Zhihui & Zhaoming & Wenjie & Bader, 2010), and computer vision (Lu & Ji, 2015;Scharstein & Pal, 2007;Delaye &Adrien, 2014;Roscher & Herzog & Kunkel, 2014;Karimaghaloo & Arnold & Arbel, 2015;Bhole &Pal, 2014). Although early applications of CRFs used linear chains, recent applications of CRFs have also used more general graphical structures.…”

Section: Figure 1 Face Image Segmentationmentioning

confidence: 99%

A Study on Components and Features in Face Detection

Yang

Wang

2015

International Journal of Information Technology and Web Engineering

View full text Add to dashboard Cite

It's natural and convenient that the face is used as a feature in the city safety monitoring. Components and feature extraction are two key problems in face detection. To address the local occlusion and pose variation in face detection, face can be looked on as a whole composed of several parts from up to down. First, the face is divided into a number of local regions from which various features are extracted. Each region is identified by a local classifier and is assigned a preliminary part label. A random field is established based on these labels and multiple dependencies between different parts are modeled in a CRF framework. The probability that the test image may be a face is calculated by a trained CRF model. The probability is used as a measure to test the existence of a face. The experiments were carried out on the CMU/MIT dataset. As indicated by the experiment results, the following methods can improve the detection rate and enhance the robustness of face detection in case of occlusion: 1) integrating multiple features and multiple dependencies in CRF framework; 2) dividing the face optimally.

show abstract

3D segmentation of abdominal CT imagery with graphical models, conditional random fields and learning

Cited by 23 publications

References 37 publications

Asymptotically efficient parameter estimation in hidden Markov spatio-temporal random fields

Asymptotically efficient parameter estimation in hidden Markov spatio-temporal random fields

Computational anatomy for multi-organ analysis in medical imaging: A review

A Study on Components and Features in Face Detection

Contact Info

Product

Resources

About