Retrospective shading correction based on entropy minimization

Likar, Boštjan; Maintz, J. B. Antoine; Viergever, Max A.; Pernuš, Franjo

doi:10.1046/j.1365-2818.2000.00669.x

Cited by 122 publications

(117 citation statements)

References 14 publications

(11 reference statements)

Supporting

Mentioning

116

Contrasting

Unclassified

Order By: Relevance

“…Finally, through a web interface such as CAIMAN, where no programming or downloads are required. Up to the authors' best knowledge, only one webpage created by Likar and co-authors [18] provides a similar approach to CAIMAN. This webpage processes images for intensity inhomogenetiy correction using an information minimization algorithm [19].…”

Section: Alternative Approaches For Algorithm Disseminationmentioning

confidence: 99%

CAIMAN: An online algorithm repository for Cancer Image Analysis

Reyes-Aldasoro

Griffiths

Savas

et al. 2011

Computer Methods and Programs in Biomedicine

View full text Add to dashboard Cite

show abstract

Section: Alternative Approaches For Algorithm Disseminationmentioning

confidence: 99%

CAIMAN: An online algorithm repository for Cancer Image Analysis

Reyes-Aldasoro

Griffiths

Savas

et al. 2011

Computer Methods and Programs in Biomedicine

View full text Add to dashboard Cite

show abstract

“…Several steps were taken to quantify both FV and PV visceral fat voxel numbers. First, automated intensity correction was performed on each slice, which iteratively applied 2D third-order polynomial field correction to find the minimal entropy of the image grayscale histogram (14). This procedure reduced the signal nonuniformity due to B 0 and B 1 inhomogeneities.…”

Section: Partial-volume Fat Quantificationmentioning

confidence: 99%

Impact of partial volume effects on visceral adipose tissue quantification using MRI

Zhou

Murillo

Peng

2011

Magnetic Resonance Imaging

View full text Add to dashboard Cite

Purpose: To quantitatively estimate the impact of partial volume effects on visceral adipose tissue (VAT) quantification using typical resolution magnetic resonance imaging (MRI). Materials and Methods:Nine normal or overweight subjects were scanned at central abdomen levels with a water-saturated, balanced steady-state free precession (b-SSFP) sequence. The water-saturation effectiveness was evaluated with region-of-interest analysis on fat, muscle, bowel, and noise areas. The number of full-volume (FV) and partial-volume (PV) fat pixels was estimated based on a gray-level histogram model of water-saturated images. Both FV and PV fat amounts were quantified.Results: High-quality, fat-only images were generated with the b-SSFP imaging method. Fat SNR was 77.7 6 25.6 and water-saturation was effective, with the average fat-to-water signal intensity ratio ¼ 20.7 6 3.8. The average ratio of partial-to full-volume fat amounts was 104.0%. The ratio was higher with lower body mass index (BMI) and PV fat amount only increased slightly when BMI increased.Conclusion: PV fat contributes a significant amount of fat to fat measurements on typical spatial resolution MRI on normal and overweight subjects. The relative PV fat contribution is markedly higher in slimmer patients. Inclusion of this portion of the adipose tissue will increase overall accuracy and decrease variability of VAT quantification using MRI. IMAGING OF HUMAN BODY FAT and its distribution have gained renewed attention because of the link between obesity and physical inactivity to multiple disease states. Excessive body fat, particularly visceral adipose tissue (VAT), correlates with adverse metabolic consequences especially those related to the metabolic syndrome such as diabetes type 2 and cardiovascular disease (1-3). Moreover, the global obesity epidemic has heightened the importance and magnitude of the problems (4). Thus, accurate and rapid methods for measuring VAT have been sought to better investigate its contribution to many disease states as well as the development of diagnostic biomarkers (5,6).VAT imaging and quantification have been challenging because of motion artifacts, the highly complicated nature of anatomic fat compartments, and the disseminated nature of visceral fat. Compared with computerized tomography (CT), magnetic resonance imaging (MRI) has no associated ionizing radiation and is more desirable for repeated studies for infants, children, and healthy subjects. MRI is therefore the method of choice for VAT quantification in such clinical research studies despite its known relatively poor spatial resolution due to signal-to-noise ratio (SNR) and scan duration limitation of a clinical scan. Human abdominal fat quantification by MRI has been historically performed using T1-weighted (T1W) turbo spin-echo (TSE) or gradient-echo imaging techniques. These techniques can generate moderate contrast between fat and nonfat tissues to enable classification of these pixels using postprocessing methods such as a manual contour drawing method (7)...

show abstract

“…Many image acquisition systems contain artifacts that are hard to correct with calibration schemes. One example in medical image processing is the inhomogeneity of the magnetic field of an MR scanner or of the sensitivity of MR surface coils, leading to low frequency gradients over the image, for which retrospective correction schemes have been proposed [156,157]. A generated example is shown in Figure 4.7 where we multiplied a texture image with Gaussian noise.…”

Section: Histogram Transformationsmentioning

confidence: 99%