&lt;title&gt;Individual 3D region-of-interest atlas of the human brain: automatic training point extraction for neural-network-based classification of brain tissue types&lt;/title&gt;

Wagenknecht, Gudrun; Kaiser, Hartmut; Obladen, Thorsten; Sabri, O.; Buell, Udalrich

doi:10.1117/12.382908

Cited by 2 publications

(8 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Especially the histogram-based segmentation of gray and white matter could be a problem since the histogram curve can show a smooth transition from GM to WM. For this reason, more sophisticated statistical and neural network approaches for classification of individual data sets were developed [2,3]. For development of special software phantoms, however, this method is sufficient.…”

Section: Classification Of Tissue Typesmentioning

confidence: 99%

“…C: XxYxZ->C (2) with CE C={O Crnax}={BG,CSF,GM,WM,SB} c(x,y,z)=C and C=f(F) Classification, expressed as a function C=J(F,), means that the assignment of each feature (vector) F to a class C is unique.…”

Section: Classification Of Tissue Typesmentioning

confidence: 99%

“…Two examples of applications of this software phantom: 1 . evaluation of a dynamic programming (DP) algorithm for contrast correction based on the cumulative grayscale value histograms of neighboring slices of gradient-spoiled T1-weighted 3D-FLASH image data [1]; 2. evaluation of the first main steps in individual region-of-interest (ROT) atlas extraction, which are the automatic detection of training points and supervised classification of T1-weighted MRI grayscale images into brain tissue types (GM, WM, CSF, SB, BG) [2,3]. Error rates for these examples are calculated based on this software phantom.…”

Section: Spoiling Artifact (30)mentioning

confidence: 99%

See 2 more Smart Citations

<title>Simulation of 3D MRI brain images for quantitative evaluation of image segmentation algorithms</title>

et al. 2000

Self Cite

View full text Add to dashboard Cite

To model the true shape of MRI brain images, automatically classified Ti-weighted 3D MRI images (gray matter, white matter, cerebrospinal fluid, scalp/bone and background) are utilized for simulation of grayscale data and imaging artifacts. For each class, Gaussian distribution of grayscale values is assumed, and mean and variance are computed from grayscale images. A random generator fills up the class images with Gauss-distributed grayscale values. Since grayscale values of neighboring voxels are not correlated, a Gaussian low-pass filtering is done, preserving class region borders. To simulate anatomical variability, a Gaussian distribution in space with user-defined mean and variance can be added at any user-defined position. Several imaging artifacts can be added: 1 . to simulate partial volume effects, every voxel is averaged with neighboring voxels if they have a different class label; 2. a linear or quadratic bias field can be added with user-defined strength and orientation; 3. additional background noise can be added; and 4. artifacts left over after spoiling can be simulated by adding a band with increasing/decreasing grayscale values. With this method, realistic-looking simulated MRI images can be produced to test classification and segmentation algorithms regarding accuracy and robustness even in the presence of artifacts.Evaluation of classification or segmentation results and validation of automatic algorithms is difficult in the case of nonsimulated MRI images because the true borders of tissue types are unknown. Therefore, evaluation is often done by comparing automatically obtained results with results supplied by a physician. This has the disadvantage of subjectivity and inter-or intra-observer variability. Furthermore, 3D brain images have very complex region borderlines. A complete interactive delineation is practically impossible.In this approach, Ti-weighted 3D-FLASH MRI grayscale images are utilized to generate a software phantom for evaluation and validation purposes. The images are classified into gray matter (GM), white matter (WM), cerebrospinal fluid (CSF), scalpfbone (SB) and background (BG) using a simple thresholding method. To model the true shape of MRI brain images, these automatically classified 3D MIRI images are utilized for simulation of grayscale data and imaging artifacts.The main topics of this paper are the detailed descriptions of the methods developed for classification and simulation of Ti-weighted 3D-MRI images and possible artifacts. Examples of applications of this software phantom are briefly described.

show abstract

Section: Classification Of Tissue Typesmentioning

confidence: 99%

Section: Classification Of Tissue Typesmentioning

confidence: 99%

Section: Spoiling Artifact (30)mentioning

confidence: 99%

See 1 more Smart Citation

<title>Simulation of 3D MRI brain images for quantitative evaluation of image segmentation algorithms</title>

et al. 2000

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this approach, uniqueness of classification is forced and classification is done without a rejection class. c s and c = u ... with 2 n =0 for: i j (4) The sample is used to adapt the classifier to the given problem. The classifier is then used to assign all other features (vectors) of the population to one of the classes C.…”

Section: Supervised Classificationmentioning

confidence: 99%

“…Individual region-of-interest (IROI) atlas extraction consists of two main parts [4][5][6]: since functional parameters are mainly determined in regions of gray and white matter, the exact voxel-based classification of these tissue types and the regions they form, as well as the differentiation from spatially neighboring tissue types, is essential. Thus, the first step in atlas extraction is voxel-based classification of T1-weighted MRI grayscale images (e.g., 3D-FLASH) into gray matter (GM), white matter (WM), cerebrospinal fluid (CSF), scalp/bone (SB), and background (BG).…”

Section: Introductionmentioning

confidence: 99%

<title>Individual 3D region-of-interest atlas of the human brain: neural-network-based tissue classification with automatic training point extraction</title>

et al. 2000

Self Cite

View full text Add to dashboard Cite

The purpose of individual 3D region-of-interest atlas extraction is to automatically define anatomically meaningful regions in 3D MRT images for quantification of functional parameters (PET, SPECT: rMRG1u, rCBF). The first step of atlas extraction is to automatically classify brain tissue types into gray matter (GM), white matter (WM), cerebrospinal fluid (CSF), scalp/bone (SB) and background (BG). A feed-forward neural network with back-propagation training algorithm is used and compared to other numerical classifiers. It can be trained by a sample from the individual patient data set in question.Classification is done by a "winner takes all' decision. Automatic extraction of a user-specified number of training points is done in a cross-sectional slice. Background separation is done by simple region growing. The most homogeneous voxels define the region for WM training point extraction (TPE). Non-white-matter and non-background regions are analyzed for GM and CSF training points. For SB TPE, the distance from the BG region is one feature. For each class, spatially uniformly distributed training points are extracted by a random generator from these regions. Simulated and real 3D MRI images are analyzed and error rates for TPE and classification calculated. The resulting class images can be analyzed for extraction of anatomical ROIs.

show abstract

<title>Individual 3D region-of-interest atlas of the human brain: automatic training point extraction for neural-network-based classification of brain tissue types</title>

Cited by 2 publications

References 0 publications

<title>Simulation of 3D MRI brain images for quantitative evaluation of image segmentation algorithms</title>

<title>Simulation of 3D MRI brain images for quantitative evaluation of image segmentation algorithms</title>

<title>Individual 3D region-of-interest atlas of the human brain: neural-network-based tissue classification with automatic training point extraction</title>

Contact Info

Product

Resources

About