Fig 7 - uploaded by Robert Lenkinski
Content may be subject to copyright.
Osteomyelitis of the thoracic spine at T6-T7. (A) and (e) are uncorrected images. (8) and (0) are corrected images. (AI and (8) are sagittal images while (C) and (0) are axial images at the level of the lesion. Note the enhanced appreciation of surrounding soft tissues.
Source publication
The increasing use of digital image data in Radiology has opened the door to the routine use of numerical image-enhancement techniques. Of course, numerical image processing cannot put information into the image which is not already there. However, if some means can be found to separate diagnostic image information from noise or artifact, the diagn...
Context in source publication
Context 1
... the authors' institution, correction has been found to increase confidence in images in which the area of interest extends across the drop-off in surface- coil response, or in situations where the radiolo- gist wants to make a comparison of tissue con- trasts in normal and suspect regions of anatomy (Fig 6). This is especially helpful in situations in which there is widespread soft tissue involvement in pathology, as in Fig 7, a case of osteomyelitis in the thoracic spine. ...
Similar publications
Objective: To propose a reproducible, user friendly and low cost method for digitization of radiographic films of all the standard sizes, focusing efforts on chest X-ray films. Materials and Methods: The focus on low cost have dictated the use of an A4 scanner with transparency adapter, as well as an optimized image stitching software that takes ad...
Citations
... This is the case in MRI when the magnetic field in the region of interest is not completely uniform. Several techniques have been proposed in order to correct for this position-dependent biases, but they mostly rely on obtaining good estimates of typical tissue intensities [5,12], or alternatively are simplistic models based on removing the lower frequency components of the Fourier decomposition of the image [6]. Our approach does not require the time consuming step of bias field correction due to the properties of phase congruency. ...
Feature detection on MR images has largely relied on intensity classification and gradient-based magnitudes. In this paper, we propose the use of phase congruency as a more robust detection method, as it is based on a multiscale intensity-invariant measure. We show the application of phase congruency for the detection of cortical sulci from T2 weighted MRI. Sulci represent important landmarks in the structural analysis of the brain, as their location and orientation provide valuable information for diagnosis and surgical planning. Results show that phase congruency outperforms previous techniques, even in the presence of intensity bias fields due to magnetic field inhomogeneity.
... This is the case in MRI when the magnetic field in the region of interest is not completely uniform. Several techniques have been proposed in order to correct for this position-dependent biases, but they mostly rely on obtaining good estimates of typical tissue intensities [5,12], or alternatively are simplistic models based on removing the lower frequency components of the Fourier decomposition of the image [6]. Our approach does not require the time consuming step of bias field correction due to the properties of phase congruency. ...
Feature detection on MR images has largely relied on intensity classification and gradient-based magnitudes. In this paper, we propose the use of phase congruency as a more robust detection method, as it is based on a multiscale intensity-invariant measure. We show the application of phase congruency for the detection of cortical sulci from T2 weighted MRI. Sulci represent important landmarks in the structural analysis of the brain, as their location and orientation provide valuable information for diagnosis and surgical planning. Results show that phase congruency outperforms previous techniques, even in the presence of intensity bias fields due to magnetic field inhomogeneity.
... Surface coil intensity normalization (15)(16)(17)(18)(19)(20) removes the large variation in image intensity due to the rapid fall-off in the surface coil field, thereby greatly improving the visualization of local tissue contrast. By using intensity normalization with phase-sensitive reconstruction, the image intensity window and level may be adjusted to maximize the contrast ratio (21), effectively shifting the null point without reacquiring additional images at various inversion times. ...
... Techniques that require separate acquisition of surface and body coil are subject to motion-related errors, and require additional breath-hold image acquisitions. Methods have been described that use only surface coil data and suppress image features by spatial blurring (15)(16)(17). Image features that are not suppressed will somewhat alter the contrast between tissue types. The noise will be greatly amplified in regions where the tissue is approximately nulled, such as normal myocardium. ...
After administration of gadolinium, infarcted myocardium exhibits delayed hyperenhancement and can be imaged using an inversion recovery (IR) sequence. The performance of such a method when using magnitude-reconstructed images is highly sensitive to the inversion recovery time (TI) selected. Using phase-sensitive reconstruction, it is possible to use a nominal value of TI, eliminate several breath-holds otherwise needed to find the precise null time for normal myocardium, and achieve a consistent contrast. Phase-sensitive detection is used to remove the background phase while preserving the sign of the desired magnetization during IR. Experimental results are presented which demonstrate the benefits of both phase-sensitive IR image reconstruction and surface coil intensity normalization for detecting myocardial infarction (MI). The phase-sensitive reconstruction method reduces the variation in apparent infarct size that is observed in the magnitude images as TI is changed. Phase-sensitive detection also has the advantage of decreasing the sensitivity to changes in tissue T(1) with increasing delay from contrast agent injection.
Detailed radiologic–pathologic correlation has served as the basis for many of the MRI applications in the abdomen, pelvis and musculoskeletal system developed by our multidisciplinary research group at the University of Pennsylvania. Following the analysis of surgical and pathologic specimens, we frequently modified MR technology to enhance visualization of the region, analyzed the relative utility of imaging findings, and finally assessed diagnostic performance using receiver operating characteristic methodology. This article describes the history of our use of this approach.
Signal inhomogeneities in volumetric head MR scans are a major obstacle to segmentation and neuromorphometry. The fuzzy c-means (FCM) statistical clustering algorithm was extended to estimate and retrospectively correct a multiplicative inhomogeneity field in T1-weighted head MR scans. The method was tested on a mathematically simulated object and on seven whole head 3D MR scans. Once initial parameters governing operation of the algorithm were chosen for this class of images, results were obtained without intervention for individual MR studies. Post-acquisition inhomogeneity correction by extended FCM clustering improved overall image uniformity and separability of gray and white matter intensities.
The segmentation of magnetic resonance images (MRI) is a challenging problem that has received an enormous amount of attention lately. Many researchers have applied various techniques however fuzzy c-means (FCM) based algorithms have produced better results compared to other methods. In this paper, we present a modified FCM algorithm for bias (also called intensity in-homogeneities) estimation and segmentation of MRI. Normally, the intensity in-homogeneities are attributed to imperfections in the radio-frequency coils or to the problems associated with the image acquisition. Our algorithm is formulated by modifying the objective function of the standard FCM and it has the advantage that it can be applied at an early stage in an automated data analysis before a tissue model is available. The proposed method can deal with the intensity in-homogeneities and Gaussian noise effectively. We have conducted extensive experimental and have compared our results with other reported methods. The results using simulated images and real MRI data show that our method provides better results compared to standard FCM-based algorithms and other modified FCM-based techniques.
MRI data sets are corrupted by multiplicative inhomogeneities, often referred to as nonuniformities or intensity variations, that hamper the use of quantitative analyses. The use of adiabatic pulses can remove the inhomogeneity effects on transmit, but coil and patient parameters still affect reception. We describe an automatic technique that not only improves the worst corruptions such as those introduced by surface coils, but also corrects typical inhomogeneities encountered in routine volume data sets such as head scans without generating additional artifact. Because the technique uses only the patient data set, the technique can be applied retrospectively to all data sets, and corrects both patient independent effects such as rf coil design, and patient dependent effects such as tissue attenuation and dielectric-induced resonances experienced in high field MRI. Patient dependent attenuation effects are also encountered in x-ray computed tomography. All of the above are examples of multiplicative inhomogeneities which result in low spatial frequency corruption of acquired volume data sets. While we concentrate on MR in the remainder of the paper, the algorithms and techniques described are directly applicable to CT as well. Following such corrections, region of interest analyses, volume histograms, and thresholding techniques are more meaningful. The value of such correction algorithms may increase dramatically with increased use of high field strength magnets and associated patient-dependent rf attenuation and resonance effects.
Magnetic resonance imaging has become an important imaging modality for the male pelvis. Its unparalleled ability to depict soft tissue structures and highlight pathology have made it the best method for determining the extent of many disease processes. This article reviews the use of MR to evaluate diseases of the prostate gland and bladder. In both, the major indication for imaging is the local staging of cancer, and MR is currently the best imaging modality. This article will discuss the critical clinical issues concerning prostate cancer and neoplasms of the bladder, and the contribution of MR imaging.
Prostate cancer is the most common cancer in the world, the most frequently diagnosed in the United States, and the second most lethal cancer in U.S. men. Earlier diagnosis implies better prognosis. However, prognosis may be dependent upon the stage of the malignancy at the time of diagnosis and implementation of appropriate therapy. Clinical staging, even with the development of serum prostate-specific antigen and other studies, has not proven to be highly accurate, particularly to identify and quantitate local disease and extension. Imaging has, in the past, also had limited success. With the development of computed tomography (CT), endorectal ultrasound, and magnetic resonance imaging (MRI), there was great expectations for improvement. However, CT and ultrasound have not been as accurate as hoped. MRI, because of its multiorientation and multiparameter abilities, has been the most definitive imaging tool for staging of local extension, yet still has limitations. The prostate capsule, the neurovascular bundles, the seminal vesicle, and other regions prone to initial attack by cancer extension can be seen exquisitely clearly by the newer approaches to MRI. Cancer extension, however, cannot be consistently identified when it is microscopic. MRI is an accurate identifier of macroscopic, even subtle macroscopic disease, but there are still limitations in its ability to diagnose all pathology.
We propose a modification of Wells et al. technique for bias field estimation and segmentation of magnetic resonance (MR) images. We show that replacing the class other, which includes all tissue not modeled explicitly by Gaussians with small variance, by a uniform probability density, and amending the expectation-maximization (EM) algorithm appropriately, gives significantly better results. We next consider the estimation and filtering of high-frequency information in MR images, comprising noise, intertissue boundaries, and within tissue microstructures. We conclude that post-filtering is preferable to the prefiltering that has been proposed previously. We observe that the performance of any segmentation algorithm, in particular that of Wells et al. (and our refinements of it) is affected substantially by the number and selection of the tissue classes that are modeled explicitly, the corresponding defining parameters and, critically, the spatial distribution of tissues in the image. We present an initial exploration to choose automatically the number of classes and the associated parameters that give the best output. This requires us to define what is meant by "best output" and for this we propose the application of minimum entropy. The methods developed have been implemented and are illustrated throughout on simulated and real data (brain and breast MR).
