MRI: Stability of three supervised segmentation techniques

Clarke, Laurence P.; Velthuizen, R.P.; Phuphanich, Surasak; Schellenberg, J.D.; Arrington, John A.; Silbiger, Martin L.

doi:10.1016/0730-725x(93)90417-c

Cited by 145 publications

(52 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Typically, the multi-variate Gaussian model ("Normal" distribution) is used. If the features are MR signal intensities from various MRI modalities (such as T1, T2, PD), then the Gaussian model assumption can be poor (Clarke et al, 1993;DeCarli et al, 1992;Schellenberg et al, 1990) : besides biological causes such as the intrinsic heterogeneity within the tissue classes that concern this paper (CSF, grey matter, white matter), the MRI acquisition artifacts also affect the intensity distributions -i.e. result in deviations from a Normal distribution (Ashburner, 2000;Kollokian, 1996;Schellenberg et al, 1990).…”

Section: Feature Space Distributionsmentioning

confidence: 99%

A fully automatic and robust brain MRI tissue classification method

Cocosco

Zijdenbos

Evans

2003

Medical Image Analysis

297

184

View full text Add to dashboard Cite

A novel, fully automatic, adaptive, robust procedure for brain tissue classification from 3D magnetic resonance head images (MRI) is described in this paper. The procedure is adaptive in that it customizes a training set, by using a "pruning" strategy, such that the classification is robust against anatomical variability and pathology. Starting from a set of samples generated from prior tissue probability maps (a "model") in a standard, brain-based coordinate system ("stereotaxic space"), the method first reduces the fraction of incorrectly labeled samples in this set by using a minimum spanning tree graph-theoretic approach. Then, the corrected set of samples is used by a supervised kNN classifier for classifying the entire 3D image. The classification procedure is robust against variability in the image quality through a non-parametric implementation: no assumptions are made about the tissue intensity distributions. The performance of this brain tissue classification procedure is demonstrated through quantitative and qualitative validation experiments on both simulated MRI data (10 subjects) and real MRI data (43 subjects). A significant improvement in output quality was observed on subjects who exhibit morphological deviations from the model due to aging and pathology.

show abstract

Section: Feature Space Distributionsmentioning

confidence: 99%

A fully automatic and robust brain MRI tissue classification method

Cocosco

Zijdenbos

Evans

2003

Medical Image Analysis

297

184

View full text Add to dashboard Cite

show abstract

“…A number of studies have demonstrated the superiority of KNN over the Parzen Window method. 8,25 Therefore, in these studies we utilized the KNN technique. However, the proposed method is equally applicable to the Parzen Window technique with appropriate modifications.…”

Section: Introductionmentioning

confidence: 99%

Implementation of High-Dimensional Feature Map for Segmentation of MR Images

Sajja²,

Narayana

2005

Ann Biomed Eng

View full text Add to dashboard Cite

A method that considerably reduces the computational and memory complexities associated with the generation of high dimensional (≥3) feature maps for image segmentation is described. The method is based on the K-nearest neighbor (KNN) classification and consists of two parts: preprocessing of feature space and fast KNN. This technique is implemented on a PC and applied for generating threeand four-dimensional feature maps for segmenting MR brain images of multiple sclerosis patients.

show abstract

“…6). The algorithm is stable (16)(17)(18)(19). It is also one of the quickest algorithms in terms of both operator input and execution…”

Section: Mapping Response In Patient With Malignant Glioma 533mentioning

confidence: 99%

“…time (16,17). The algorithm detects anatomically relevant structures in cases where more automated algorithms have more difficulty (11,18).…”

Section: S M Haney Et Al 534mentioning

confidence: 99%

Mapping Therapeutic Response in a Patient with Malignant Glioma

Haney

Thompson²,

Cloughesy³

et al. 2001

Journal of Computer Assisted Tomography

View full text Add to dashboard Cite

Short-interval scanning of patients offers a detailed understanding of the natural progression of tumor tissue, as revealed through imaging markers such as contrast enhancement and edema, prior to therapy. Following treatment, short-interval scanning can also provide evidence of attenuation of growth rates. We present a longitudinal imaging study of a patient with glioblastoma multiforme (GBM) scanned 15 times in 104 days on a 3 T MR scanner. Images were analyzed independently by two automated algorithms capable of creating detailed maps of tumor changes as well as volumetric analysis. The algorithms, a nearest-neighbor-based tissue segmentation and a surface-modeling algorithm, tracked the patient's response to temozolomide, showing an attenuation of growth. The need for surrogate imaging end-points, of which growth rates are an example, is discussed. Further, the strengths of these algorithms, the insight gained by short-interval scanning, and the need for a better understanding of imaging markers are also described.

show abstract

MRI: Stability of three supervised segmentation techniques

Cited by 145 publications

References 19 publications

A fully automatic and robust brain MRI tissue classification method

A fully automatic and robust brain MRI tissue classification method

Implementation of High-Dimensional Feature Map for Segmentation of MR Images

Mapping Therapeutic Response in a Patient with Malignant Glioma

Contact Info

Product

Resources

About