Local histogram correction of MRI spatially dependent image pixel intensity nonuniformity

We describe a computationally straightforward post-hoc statistical method of correcting spatially dependent image pixel intensity nonuniformity based on differences in local tissue intensity distributions. Pixel intensity domains for the various tissues of the composite image are identified and compared to the distributions of local samples. The nonuniformity correction is calculated as the difference of the local sample median from the composite sample median for the tissue class most represented by the sample. The median was chosen to reduce the effect ers on determining the sample statistic and to allow a sample size small enough to accurately estimate the spatial variance of the image intensity nonuniformity. The method was designed for application to two-dimensional images. Simulations were used to estimate optimal conditions of local histogram kernel size and to test the accuracy of the method under known spatially dependent nonuniformities. The method was also applied to correct a phantom image and cerebral MRIs from 15 healthy subjects. Results show that the method accurately models simulated spatially dependent image intensity differences. Further analysis of clinical MR data showed that the variance of pixel intensities within the cerebral MRI slices and the variance of slice volumes within individuals were significantly reduced after nonuniformity correction. Improved brain-cerebrospinal fluid segmentation was also obtained. The method significantly reduced the variance of slice volumes within individuals, whether it was applied to the native images or images edited to remove nonbrain tissues. This statistical method was well behaved under the assumptions and the images tested. The general utility of the method was not determined, but conditions for testing the method under a variety of imaging sequences is discussed. We believe that this algorithm can serve as a method for improving MR image segmentation for clinical and research applications.     

MRI intensity nonuniformity correction using simultaneously spatial and gray-level histogram information.

A novel approach for correcting intensity nonuniformity in magnetic resonance imaging (MRI) is presented. This approach is based on the simultaneous use of spatial and gray-level histogram information. Spatial information about intensity nonuniformity is obtained using cubic B-spline smoothing. Gray-level histogram information of the image corrupted by intensity nonuniformity is exploited from a frequential point of view. The proposed correction method is illustrated using both physical phantom and human brain images. The results are consistent with theoretical prediction, and demonstrate a new way of dealing with intensity nonuniformity problems. They are all the more significant as the ground truth on intensity nonuniformity is unknown in clinical images.     

A MnCl2-based MR signal intensity linear response phantom

PURPOSE: This study aimed to develop a manganese chloride (MnCl2)-based phantom model that would allow progressive quantitative assessment of tissue hydration based on observed magnetic resonance (MR) imaging signal intensity (SI) linearity characteristics. MATERIALS AND METHODS: The study was performed using a progressive signal refinement technique that allowed development of an imaging tool for semiquantitative sequential discrimination of MR signal responses. A series of 82 phantoms comprising a gelatin-set MnCl2 composite were imaged under basic T1- and T2-weighted conditions. MR SI measurements were taken using region-of-interest selection, and MnCl2 concentrations were adjusted to allow development of a pair of 8-tube phantoms. These phantoms permitted progressive incremental assessment of hydration based on fundamental MR SI response. RESULTS: Statistical analysis showed that phantom MR signal response linearity can be achieved using the phantoms described under both T1 and T2 imaging conditions, yielding R2 values of 0.97 and 0.94, respectively. CONCLUSION: This novel MnCl2-based phantom can be used as a noninvasive reference standard for quantitative classification of in vivo tissue hydration based on routine clinical MR imaging sequences. Progressive correlation testing using a human cartilage sample should be performed to further refine the model for clinical application.     

Correction of B1 inhomogeneities using echo-planar imaging of water

Reliable interpretation of the MR signal intensity over the FOV of an image must consider the spatial heterogeneity of instrumental sensitivity. A major source of such variation is the nonuniformity of the B₁ magnetic field of the radiofrequency coil. This heterogeneity can be minimized by coil design but is exaggerated by surface coils, which are used to maximize the signal-to-noise ratio for some applications. This paper describes a rapid method for mapping the B₁ field over the sample of interest, using ¹H echo-planar imaging, to correct for B₁ distortions. The method applies to ¹H imaging and has been extended to non-¹H imaging by using dual-frequency coils in which the B., distributions are matched for the ¹H frequency and the frequency of interest. The approach is demonstrated in phantoms, animals, and humans and for sodium imaging.     

Contrast media for MR imaging of the heart

The role of contrast media for quantitative characterization of ischemic myocardial events with magnetic resonance (MR) imaging has advanced considerably in the past few years. Contrast material-enhanced MR imaging is useful for identifying and sizing myocardial infarcts and for distinguishing between occlusive and reperfused myocardial infarcts. Recent results suggest that contrast-enhanced MR imaging can also be used to identify areas of cell death in regions of reperfused myocardial infarction. With the aid of MR contrast media, fast MR imaging techniques may be useful in estimating regional myocardial perfusion. Although no simple relationship between signal intensity and concentration exists, contrast-enhanced MR perfusion imaging can demonstrate the presence and relative severity of hypoperfused myocardium. Combining myocardial perfusion imaging with the anatomic and functional information provided by other MR imaging techniques could make MR imaging a comprehensive noninvasive means of evaluating ischemic cardiac disease.     

The fast automatic algorithm for correction of MR bias field

PURPOSE: To develop a method for efficient automatic correction of slow-varying nonuniformity in MR images. MATERIALS AND METHODS: The original MR image is represented by a piecewise constant function, and the bias (nonuniformity) field of an MR image is modeled as multiplicative and slow varying, which permits to approximate it with a low-order polynomial basis in a "log-domain." The basis coefficients are determined by comparing partial derivatives of the modeled bias field with the original image. RESULTS: We tested the resulting algorithm named derivative surface fitting (dsf) on simulated images and phantom and real data. A single iteration was sufficient in most cases to produce a significant improvement to the MR image's visual quality. dsf does not require prior knowledge of intensity distribution and was successfully used on brain and chest images. Due to its design, dsf can be applied to images of any modality that can be approximated as piecewise constant with a multiplicative bias field. CONCLUSION: The resulting algorithm appears to be an efficient method for fast correction of slow varying nonuniformity in MR images.     

Treatment assessment of radiotherapy using MR functional quantitative imaging

Recent developments in magnetic resonance(MR) functional quantitative imaging have made it a potentially powerful tool to assess treatment response in radiation therapy. With its abilities to capture functional information on underlying tissue characteristics, MR functional quantitative imaging can be valuable in assessing treatment response and as such to optimize therapeutic outcome. Various MR quantitative imaging techniques, including diffusion weighted imaging, diffusion tensor imaging, MR spectroscopy and dynamic contrastenhanced imaging, have been investigated and found useful for assessment of radiotherapy. However, various aspects including data reproducibility, interpretation of biomarkers, image quality and data analysis impose challenges on applications of MR functional quantitative imaging in radiotherapy assessment. All of these challenging issues shall be addressed to help us understand whether MR functional quantitative imaging is truly beneficial and contributes to future development of radiotherapy. It is evident that individualized therapy is the future direction of patient care. MR functional quantitative imaging might serves as an indispensable tool towards this promising direction.     

Fat suppression in MR imaging: techniques and pitfalls.

Fat suppression is commonly used in magnetic resonance (MR) imaging to suppress the signal from adipose tissue or detect adipose tissue. Fat suppression can be achieved with three methods: fat saturation, inversion-recovery imaging, and opposed-phase imaging. Selection of a fat suppression technique should depend on the purpose of the fat suppression (contrast enhancement vs tissue characterization) and the amount of fat in the tissue being studied. Fat saturation is recommended for suppression of signal from large amounts of fat and reliable acquisition of contrast material-enhanced images. The main drawbacks of this technique are sensitivity to magnetic field nonuniformity, misregistration artifacts, and unreliability when used with low-field-strength magnets. Inversion-recovery imaging allows homogeneous and global fat suppression and can be used with low-field-strength magnets. However, this technique is not specific for fat, and the signal intensity of tissue with a long T1 and tissue with a short T1 may be ambiguous. Opposed-phase imaging is a fast and readily available technique. This method is recommended for demonstration of lesions that contain small amounts of fat. The main drawback of opposed-phase imaging is unreliability in the detection of small tumors embedded in fatty tissue.     

An improved level set method for brain MR images segmentation and bias correction

Intensity inhomogeneities cause considerable difficulty in the quantitative analysis of magnetic resonance (MR) images. Thus, bias field estimation is a necessary step before quantitative analysis of MR data can be undertaken. This paper presents a variational level set approach to bias correction and segmentation for images with intensity inhomogeneities. Our method is based on an observation that intensities in a relatively small local region are separable, despite of the inseparability of the intensities in the whole image caused by the overall intensity inhomogeneity. We first define a localized K-means-type clustering objective function for image intensities in a neighborhood around each point. The cluster centers in this objective function have a multiplicative factor that estimates the bias within the neighborhood. The objective function is then integrated over the entire domain to define the data term into the level set framework. Our method is able to capture bias of quite general profiles. Moreover, it is robust to initialization, and thereby allows fully automated applications. The proposed method has been used for images of various modalities with promising results.     

Small renal masses: assessment of lesion characterization and vascularity on dynamic contrast-enhanced MR imaging with fat suppression

OBJECTIVE: The aim of our study was to characterize renal lesions equal to or smaller than 3.0 cm using dynamic contrast-enhanced MR imaging with fat suppression by means of quantitative analysis of signal intensity. MATERIALS AND METHODS: We retrospectively reviewed the MR imaging examinations of 35 patients (20 with renal cell carcinoma, eight with angiomyolipoma, and seven with complicated cysts) who were studied with spin-echo and dynamic fat-suppressed gradient-recalled echo MR sequences, before and after the administration of gadopentetate dimeglumine. Every 30 sec after contrast injection, we measured the lesion percentage of enhancement and the ratio of contrast (lesion-renal cortex signal intensity difference) to noise. RESULTS: Ten renal cell carcinomas were classified as hypervascular (enhancement greater than that of renal cortex) and 10 as hypovascular. The percentage of enhancement of hypervascular carcinomas was similar to that of renal cortex until 150 sec and greater in the late sequences (180-210 sec, p < 0.01). Hypovascular carcinomas had a lower percentage of enhancement than hypervascular carcinomas (60-210 sec, p < 0.005). Angiomyolipomas, after an early enhancement peak, showed values similar to those of hypovascular carcinomas. Complicated cysts had very low enhancement (p < 0.001). The baseline contrast-to-noise ratio was negative for all lesions (hypointensity with respect to renal cortex). After gadolinium injection, the contrast-to-noise ratio of hypervascular carcinomas rose, becoming positive after 150 sec. Until 60 sec, the contrast-to-noise ratio of hypovascular carcinomas declined slightly, whereas that of angiomyolipomas and cysts fell sharply; then the three curves remained stable (60-210 sec, p < 0.05 for all matches except angiomyolipomas versus cysts). CONCLUSION: Quantitative analysis of signal intensity variations during dynamic contrast-enhanced MR imaging with fat suppression can be useful in the characterization of small renal lesions.     

Measurement and correction of transmitter and receiver induced nonuniformities in vivo.

Signal intensity nonuniformities in high field MR imaging limit the ability of MRI to provide quantitative information and can negatively impact diagnostic scan quality. In this paper, a simple method is described for correcting these effects based on in vivo measurement of the transmission field B1+ and reception sensitivity maps. These maps can be obtained in vivo with either gradient echo (GE) or spin echo (SE) imaging sequences, but the SE approach exhibits an advantage over the GE approach for correcting images over a range of flip angles. In a uniform phantom, this approach reduced the ratio of the signal SD to its mean from around 30% before correction to approximately 6% for the SE approach and 9% for the GE approach after correction. The application of the SE approach for correcting intensity nonuniformities is demonstrated in vivo with human brain images obtained using a conventional spin echo sequence at 3.0 T. Furthermore, it is also shown that this in vivo B1+ and reception sensitivity mapping can be performed using segmented echo planar imaging sequences providing acquisition times of less than 2 min. Although the correction presented here is demonstrated with a simultaneous transmit and receive volume coil, it can be extended to the case of separate transmission and reception coils, including surface and phase array coils.     

Fatty replacement of spinal bone marrow due to radiation: demonstration by dual energy quantitative CT and MR imaging

Dual energy CT and quantitative magnetic resonance (MR) imaging were used to evaluate marrow changes due to radiation. The bright signal intensity seen on MR was shown by the two quantitative techniques to be due to a threefold increase in the marrow fat content compared with nonradiated levels and to a normal control. Fat estimates by MR and dual energy CT were in excellent agreement. Single energy CT overestimates the amount of bone loss in the radiation field. Dual energy CT and quantitative MR can be used to correct this error.     

Intrarenal kinetics of Gd-DOTA in sequential MRI in rabbits. Reproducibility study

Ten normal rabbits and seven rabbits with experimental acute renal failure by tubular necrosis were studied with dynamic MR to evaluate the reproducibility of intrarenal kinetics of Gd-DOTA. Sequential spin-echo sequences with short TR (200 msec)/TE (26 msec) were used yielding a 29 sec acquisition time. A usual semi-quantitative analysis of intrarenal contrast demonstrated the reproducibility of some phases of the dynamic sequence in particular a drop in the signal within inner medulla between the third and the fourth minute after infusion. This effect, related to a high concentration of Gd-DOTA within the tubules was observed in 9 over 10 normal rabbits and in none of the rabbits with acute renal failure. The quantitative analysis calculation was based on relative signal intensity and contrast-to-noise ratio from the absolute signal intensity measure on regions-of-interest (ROI) on the cortex, outer medulla and inner medulla. No reproducibility of the variations with time of these parameters could be assessed. A great number of factors of variations or error, mainly during the measurements of signal intensity with ROI, could explain this lack of reproducibility. At the present, dynamic MR is therefore not able to quantitatively evaluate the renal function. Only a semi-quantitative estimation of tubular concentration can be deduced.     

An index system for comparative parameter weighting in MR imaging

An analytic method for comparative parameter weighting in magnetic resonance (MR) imaging has been developed using the concept of "fractional sensitivity." This new approach results in easily calculated indexes for T1, T2, and hydrogen weighting. This index system enables quantitative comparisons to be made between MR studies that have been performed at various field strengths, using different pulse sequences and pulse timing intervals.     

Quantitative analysis of MR images in asphyxiated neonates: correlation with neurodevelopmental outcome

BACKGROUND AND PURPOSE: MR imaging has been shown to be of prognostic significance in the evaluation of asphyxiated neonates. The purpose of this project was to determine whether the use of intensity ratios in key regions of the brain might better detect regions of injured brain and thus improve the correlation of imaging findings with 12-month neurodevelopmental outcome. METHODS: Prospectively acquired MR studies of 53 asphyxiated neonates were reviewed retrospectively. Signal intensities from standard T1- and T2-weighted images of seven major brain regions that are affected in asphyxia were measured. Intensity ratios were calculated by dividing the signal intensity of each brain region by the signal intensity of the ocular vitreous. The intensity ratios were then correlated with 12-month neurodevelopmental outcome. These results were compared with correlations determined by a qualitative scoring system. RESULTS: The only significant statistical correlation between the intensity ratios and 12-month neurodevelopmental outcome were those of anterior watershed injury with the Mental Development Index of the Bayley Scales of Infant Development II. The qualitative measurements showed a strong correlation with many outcome parameters. CONCLUSION: Standard qualitative assessment is more predictive of neurodevelopmental outcome than is quantitative analysis. This finding most likely reflects the inability of the quantitative assessment of intensity ratios to compensate for the day-to-day evolution of signal intensity of the injured neonatal brain. Anterior watershed injury may be predictive of abnormal cognitive outcome; examination of these patients at age 30 months will be important to determine the accuracy of this observation.     

Magnetic resonance first pass perfusion imaging for detecting coronary artery disease

Magnetic resonance first pass perfusion imaging can be used to detect abnormalities in myocardial blood flow. This technique involves imaging the first pass of gadolinium based contrast through the myocardium. Images are initially read qualitatively for areas of reduced signal intensity. Additionally, at our institution a quantitative method is applied that can aid both detection and diagnosis of perfusion defects. This method involves fitting the myocardial signal intensity curves and then calculates absolute myocardial blood flow. Our approach to first pass perfusion imaging will be reviewed. Magnetic resonance first pass perfusion imaging has a complimentary role with coronary angiography either non-invasively using CT or with catheterization. Perfusion imaging defines the physiology and angiography in the anatomy of coronary artery disease.     

Soft tissue tumors: post-treatment imaging

As the radiologic evaluation of soft tissue masses has changed dramatically with the advent of MR imaging, the effect of MR imaging is even more striking in the assessment of patients after treatment. In cases of local tumor recurrence, MR imaging has become the standard of care. Using a few basic principles, even small local recurrences can be detected accurately, and recurrence can be distinguished from postoperative or post-treatment change. This review presents a fundamental approach to the evaluation of patients, following treatment for soft tissue tumors and highlighting MR imaging.     

Method for cortical bone structural analysis from magnetic resonance images

RATIONALE AND OBJECTIVES: Quantitative evaluation of cortical bone architecture as a means to assess bone strength typically is accomplished on the basis of images obtained by means of dual-energy X-ray absorptiometry (DXA) or computed tomography. Magnetic resonance (MR) imaging has potential advantages for this task in that it allows imaging in arbitrary scan planes at high spatial resolution. However, several hurdles have to be overcome to make this approach practical, including resolution of issues related to nonlinear receive coil sensitivity, variations in marrow composition, and the presence of periosteal isointense tissues, which all complicate segmentation. The aim of this study is to develop MR acquisition and analysis methods optimized for the detection of cortical boundaries in such complex geometries as the femoral neck. MATERIALS AND METHODS: Cortical boundary detection is achieved by radially tracing intensity profiles that intersect the periosteal and endosteal boundaries of bone. Profiles subsequently are normalized to the intensity of the marrow signal, processed with morphologic image operators, and binarized. The resulting boundaries are mapped back onto the spatial image, and erroneous boundary points are removed. From the detected cortical boundaries, cortical cross-sectional area and thickness are computed. The method was evaluated on cortical bone specimens and human volunteers on the basis of high-resolution images acquired at a 1.5-Tesla field strength. To assess whether the method is sensitive to detect the expected dependencies of cortical parameters in weight-bearing bone on overall habitus, 10 women aged 46-73 years (mean age, 56 years) underwent the cortical imaging protocol in the proximal femur, and results were compared with DXA bone mineral density parameters of the hip and spine. RESULTS: Reproducibility was approximately 2%. Double oblique images of the femoral neck in the 10 women studied showed that cortical cross-sectional area correlated strongly with height (r = 0.88; p = .0008), whereas cortical diameter versus age approached significance (r = 0.61; p = .06). Measurements in specimens of some cortical parameters indicated resolution dependence. However, note that specimen ranking within each parameter remained constant across all resolutions studied. CONCLUSION: Data suggest the new method to be robust and applicable on standard clinical MR scanners at arbitrary anatomic locations to yield clinically meaningful quantitative results.     

Quantitative morphologic evaluation of magnetic resonance imaging during and after treatment of childhood leukemia

Introduction Medical advances over the last several decades, including CNS prophylaxis, have greatly increased survival in children with leukemia. As survival rates have increased, clinicians and scientists have been afforded the opportunity to further develop treatments to improve the quality of life of survivors by minimizing the long-term adverse effects. When evaluating the effect of antileukemia therapy on the developing brain, magnetic resonance (MR) imaging has been the preferred modality because it quantifies morphologic changes objectively and noninvasively. Method and results Computer-aided detection of changes on neuroimages enables us to objectively differentiate leukoencephalopathy from normal maturation of the developing brain. Quantitative tissue segmentation algorithms and relaxometry measures have been used to determine the prevalence, extent, and intensity of white matter changes that occur during therapy. More recently, diffusion tensor imaging has been used to quantify microstructural changes in the integrity of the white matter fiber tracts. MR perfusion imaging can be used to noninvasively monitor vascular changes during therapy. Changes in quantitative MR measures have been associated, to some degree, with changes in neurocognitive function during and after treatment. Conclusion In this review, we present recent advances in quantitative evaluation of MR imaging and discuss how these methods hold the promise to further elucidate the pathophysiologic effects of treatment for childhood leukemia.     

Signal intensity artifacts in clinical MR imaging.

Signal intensity artifacts are often encountered during magnetic resonance (MR) imaging. Occasionally, these artifacts are severe enough to degrade image quality and interfere with interpretation. Signal intensity artifacts inherent in local coil imaging include intensity gradients and local intensity shift artifact. The latter can be minimized but not eliminated with optimal coil design and tuning. Improper coil or patient positioning can produce subtle or, in some cases, severe signal intensity artifacts, and each is easily corrected. Signal intensity artifacts and image degradation can also occur in a perfectly functioning coil if protocols are not optimized. Failure of decoupling mechanisms can produce signal intensity artifacts that will not respond to protocol optimization and will worsen with gradient imaging. Improper coil tuning manifests as a shading artifact that can mimic other findings. Signal-degrading artifacts may be caused by a ferromagnetic foreign body in the imager. Signal intensity artifacts can also result from performing ultrafast imaging with coils that were not designed for this type of imaging or from MR imaging system malfunction. Familiarity with the various causes of signal intensity artifacts is necessary to maintain optimal image quality and should be required as part of any MR imaging quality assurance program.