Regression forests for efficient anatomy detection and localization in computed tomography scans

Criminisi, Antonio; Robertson, Duncan; Konukoğlu, Ender; Shotton, Jamie; Pathak, Dev S.; White, Steven; Siddiqui, Khan M.

doi:10.1016/j.media.2013.01.001

Cited by 217 publications

(210 citation statements)

References 18 publications

Supporting

Mentioning

205

Contrasting

Order By: Relevance

“…We used 10 random scans from the training data to characterize the centroid coordinates of the biomarkers with long-range feature boxes following Ref. 18 and yielded the estimated biomarkers positions on the testing data with a mean distance error of 14.43 mm. Four bounding positions were empirically defined among the vertical position of the three biomarkers to evenly distribute the available training data (25, 35, 50, 31, 36 slices for each class, ordered from bottom to top).…”

Section: Methodsmentioning

confidence: 99%

Abdomen and spinal cord segmentation with augmented active shape models

Conrad

Baucom

et al. 2016

J. Med. Imag

View full text Add to dashboard Cite

Abstract. Active shape models (ASMs) have been widely used for extracting human anatomies in medical images given their capability for shape regularization of topology preservation. However, sensitivity to model initialization and local correspondence search often undermines their performances, especially around highly variable contexts in computed-tomography (CT) and magnetic resonance (MR) images. In this study, we propose an augmented ASM (AASM) by integrating the multiatlas label fusion (MALF) and level set (LS) techniques into the traditional ASM framework. Using AASM, landmark updates are optimized globally via a region-based LS evolution applied on the probability map generated from MALF. This augmentation effectively extends the searching range of correspondent landmarks while reducing sensitivity to the image contexts and improves the segmentation robustness. We propose the AASM framework as a two-dimensional segmentation technique targeting structures with one axis of regularity. We apply AASM approach to abdomen CT and spinal cord (SC) MR segmentation challenges. On 20 CT scans, the AASM segmentation of the whole abdominal wall enables the subcutaneous/visceral fat measurement, with high correlation to the measurement derived from manual segmentation. On 28 3T MR scans, AASM yields better performances than other state-of-the-art approaches in segmenting white/gray matter in SC.

show abstract

Section: Methodsmentioning

confidence: 99%

Abdomen and spinal cord segmentation with augmented active shape models

Conrad

Baucom

et al. 2016

J. Med. Imag

View full text Add to dashboard Cite

show abstract

“…[Reynolds et al 2011], [Shotton et al 2012], [Criminisi et al 2013]). A brief introduction to the traditional Random Forest algorithm below is followed by our modifications and the reasons for introducing them.…”

Section: Before Alignment After Alignmentmentioning

confidence: 99%

“…While Random Forests in general have been well-explored already, their use for practical multivariate regression has been limited [Criminisi et al 2013]. One of the challenges lies in computing node impurity-in classification, this can be done easily by counting samples belonging to each class, whereas in regression, one needs to evaluate the probability density function, which can be costly in high dimensions.…”

Section: Before Alignment After Alignmentmentioning

confidence: 99%

Learning to Remove Soft Shadows

2015

View full text Add to dashboard Cite

Manipulated images lose believability if the user's edits fail to account for shadows. We propose a method that makes removal and editing of soft shadows easy. Soft shadows are ubiquitous, but remain notoriously difficult to extract and manipulate. We posit that soft shadows can be segmented, and therefore edited, by learning a mapping function for image patches that generates shadow mattes. We validate this premise by removing soft shadows from photographs with only a small amount of user input.Given only broad user brush strokes that indicate the region to be processed, our new supervised regression algorithm automatically unshadows an image, removing the umbra and penumbra. The resulting lit image is frequently perceived as a believable shadow-free version of the scene. We tested the approach on a large set of soft shadow images, and performed a user study that compared our method to the state of the art and to real lit scenes. Our results are more difficult to identify as being altered, and are perceived as preferable compared to prior work.

show abstract

“…It comprised the regression of a 6 Dimensional (6D) displacement vector (offset vector) from the bounding box walls of the organs to any given voxel. In 2013, the same authors proposed a modified implementation of RRFs [3,4] that enhanced the previously achieved results by modifying the split node optimization method, the description of the random process, and the eventual usage of this description for prediction.…”

Section: Introductionmentioning

confidence: 99%

“…Given a set of voxels arriving at a particular node and for a given organ, the current setting of RRF regresses the continuous conditional distribution of d(v; c) as a 6D multivariate Gaussian [2,3,4]. Consequently, the 6D multivariate Gaussian results in 6 1D univariate Gaussians.…”

Section: Introductionmentioning

confidence: 99%

Light Random Regression Forests for automatic multi-organ localization in CT images

Samarakoon

Promayon

Fouard

2017

2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017)

View full text Add to dashboard Cite

Classic Random Regression Forests (RRFs) used for multiorgan localization describe the random process of multivariate regression by storing the histograms of offset vectors along each bounding wall direction per leaf node. On the one hand, the RAM and storage requirements of classic RRFs may become exorbitantly high when such a RRF consists of many leaf nodes, but on the other hand, a large number of leaf nodes are required for better localization. We introduce Light Random Regression Forests (LRRFs) which eliminate the need to describe the random process by formulating the localization prediction based on the random variables that describe the random process. Consequently, LRRFs with the same localization capabilities require less RAM and storage space compared to classic RRFs. LRRF comprising 4 trees with 17 decision levels is approximately 9 times faster, takes 10 times less RAM, and uses 30 times less storage space compared to a similar classic RRF.

show abstract

Regression forests for efficient anatomy detection and localization in computed tomography scans

Cited by 217 publications

References 18 publications

Abdomen and spinal cord segmentation with augmented active shape models

Abdomen and spinal cord segmentation with augmented active shape models

Learning to Remove Soft Shadows

Light Random Regression Forests for automatic multi-organ localization in CT images

Contact Info

Product

Resources

About