A reproducible method for automated extraction of brain volumes from 3D human head MR images

PURPOSE: To assess the frequency and prognosis of skin recurrences after breast-conserving therapy (BCT) compared with other breast recurrences. MATERIALS AND METHODS: From 1968 to 1986, 1,624 patients with unilateral stage I or II breast cancer treated with BCT at the Joint Center for Radiation Therapy (Boston, MA) underwent gross tumor excision and received a dose of > or = 60 Gy to the tumor bed. Skin recurrences (SR) were defined as breast recurrences without associated parenchymal disease. An invasive breast recurrence with any parenchymal disease noted clinically or radiographically was scored as an other breast recurrence (OBR). Median follow-up for survivors was 137 months. RESULTS: SR represented 8% (18 of 229) of all breast recurrences and occurred in 1.1% of all patients. The outcome after local recurrence was different for patients with SR and invasive OBR. Patients with SR more frequently had uncontrolled local failure (50%; 9 of 18) than did patients with OBR (14%; 26 of 188) (P = .0007). Forty-four percent (8 of 18) of patients with SR had distant metastasis simultaneously or within 2 months of the recurrence compared with 5% (9 of 188) of invasive OBR patients (P < .0001). For patients without distant metastasis at the time of recurrence, the 5-year actuarial rate of development of distant metastasis was 60% for SR patients compared with 39% for invasive OBR patients (P = .07), and the corresponding 5-year actuarial survival rates beyond the time of local failure were 51% and 79%, respectively (P = .06). CONCLUSION: In contrast to other types of invasive breast recurrence after breast-conserving therapy, skin recurrences are rare and are associated with a significantly higher rate of distant metastasis and uncontrolled local disease as well as a lower rate of survival.     

Markov random field segmentation of brain MR images

We describe a fully-automatic three-dimensional (3-D)-segmentation technique for brain magnetic resonance (MR) images. By means of Markov random fields (MRF's) the segmentation algorithm captures three features that are of special importance for MR images, i.e., nonparametric distributions of tissue intensities, neighborhood correlations, and signal inhomogeneities. Detailed simulations and real MR images demonstrate the performance of the segmentation algorithm. In particular, the impact of noise, inhomogeneity, smoothing, and structure thickness are analyzed quantitatively. Even single-echo MR images are well classified into gray matter, white matter, cerebrospinal fluid, scalp-bone, and background. A simulated annealing and an iterated conditional modes implementation are presented.     

Unsupervised connectivity-based thresholding segmentation of midsagittal brain MR images

In this paper, we propose an algorithm for automated segmentation of midsagittal brain MR images. First, we apply thresholding to obtain binary images. From the binary images, we locate some landmarks. Based on the landmarks and anatomical information, we preprocess the binary images, which substantially simplifies the subsequent operations. To separate regions what are incorrectly merged after this initial segmentation, a new connectivity-based threshold algorithm is proposed. Assuming that some prior information about the general shape and location of objects is available, the algorithm finds a boundary between two regions using the path connection algorithm and changing the threshold adaptively. In order to test the robustness of the proposed algorithm we applied the algorithm to 120 midsagittal brain images and obtained satisfactory results.     

MR images of mycobacterial infection

With the progress of imaging technologies such as CT and MR imaging, we can obtain various informations from CT and MR images. Especially, thin-section high resolution CT (HRCT) provides very useful information for the diagnosis of mycobacterial infection and other diseases of pulmonary parenchyma. Advantages of MR images over HRCT are higher tissue contrast and multidirection capability. The T2-weighted images and Gd-DTPA enhanced T1-weighted images accurately reflect pathologic structure of the lesion. Therefore, MR images can add many useful findings to CT on the selected cases which are differentiation of tubercloma from lung cancer, chronic empyema from mesothelioma, aspergiloma from lung cancer and lymphadenopathy from lymph node metastasis. We describe the usefulness of MR images for diagnosing mycobacterial infection and its differentiation from other pulmonary diseases.     

Fast 3D registration of MR brain images using the projection correlation registration algorithm

OBJECTIVE: To determine the proportion of human sperm that respond to progesterone and to determine their capacitation state. DESIGN: The sperm population was separated according to motility by means of a Percoll density gradient; three subpopulations of low, medium, and high motility were obtained. SETTING: University-based laboratory. PARTICIPANT(S): Sperm samples from healthy donors with normal spermatogram values were used. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): The percentage of viable spermatozoa that increased the intracellular [Ca2+]i in response to progesterone was determined with a fluorescence-activated cell sorter (FACS). The percentage of capacitated spermatozoa was determined as the difference between with and without phorbol 12-myristate 13-acetate stimulus according to fluorescence microscopy and the FACS methods. RESULT(S): A significant linear relationship between the proportion of motile cells and the percentage of sperm that increases the [Ca2+]i in response to progesterone was observed with or without previous phorbol 12-myristate 13-acetate treatment. The slope of the correlation equation corresponding to the phorbol 12-myristate 13-acetate treatment was significantly lower. In addition, a significant correlation between capacitation and motility was observed. CONCLUSION(S): It seems likely that the proportion of capacitated and progesterone-responding human sperm correlates with motility.     

A simplified method to measure the diffusion tensor from seven MR images

Analytical expressions of the diffusion tensor of water, D, and of scalar invariants derived from it, are given in terms of the intensities of seven diffusion-weighted images (DWIs). These formulas simplify the post-processing steps required in diffusion tensor imaging, including estimating D in each voxel (from the set of b-matrices and their corresponding DWIs), and then computing its eigenvalues, eigenvectors, and scalar invariants. In a study conducted using artifact-free DWIs with high diffusion weighting (bmax approximately 900 s/mm2, maps of Trace(D) and the Relative and Lattice Anisotropy indices calculated analytically and by multivariate linear regression showed excellent agreement in brain parenchyma of a healthy living cat. However, the quality of the analytically computed maps degraded markedly as diffusion weighting was reduced. Although diffusion tensor MRI with seven DWIs may be useful for clinical applications where rapid scanning and data processing are required, it does not provide estimates of the uncertainty of the measured imaging parameters, rendering it susceptible to noise and systematic artifacts. Therefore, care should be taken when using this technique in radiological applications.     

Enhancement of MR images using registration for signal averaging

Recent studies indicate that arsenic may generate reactive oxygen species to exert its toxicity. However, the mechanism is still unclear. In this study, we demonstrate that arsenite is able to induce apoptosis in a concentration- and time-dependent manner; however, arsenate is unable to do so. An increase of intracellular peroxide levels was accompanied with arsenite-induced apoptosis, as demonstrated by flow cytometry using DCFH-DA. N-Acetyl-L-cysteine (a thiol-containing antioxidant), diphenylene iodonium (an inhibitor of NADPH oxidase), 4,5-dihydro-1,3-benzene disulfonic acid (a selective scavenger of O2-), and catalase significantly inhibit arsenite-induced apoptosis and intracellular fluorescence intensity. In contrast, allopurinol (an inhibitor of xanthine oxidase), indomethacin (an inhibitor of cyclooxygenase), superoxide dismutase, or PDTC had no effect on arsenite-induced cell death. Activation of CPP32 activity, PARP (a DNA repair enzyme) degradation, and release of cytochrome c from mitochondria to the cytosol are involved in arsenite-induced apoptosis, and Bcl-2 antagonize arsenite-induced apoptosis by a mechanism that interferes in the activity of CPP32. These results lead to a working hypothesis that arsenite-induced apoptosis is triggered by the generation of hydrogen peroxide through activation of flavoprotein-dependent superoxide-producing enzymes (such as NADPH oxidase), and hydrogen peroxide might play a role as a mediator to induce apoptosis through release of cytochrome c to cytosol, activation of CPP32 protease, and PARP degradation.     

Visual assessment of the accuracy of retrospective registration of MR and CT images of the brain

In a previous study we demonstrated that automatic retrospective registration algorithms can frequently register magnetic resonance (MR) and computed tomography (CT) images of the brain with an accuracy of better than 2 mm, but in that same study we found that such algorithms sometimes fail, leading to errors of 6 mm or more. Before these algorithms can be used routinely in the clinic, methods must be provided for distinguishing between registration solutions that are clinically satisfactory and those that are not. One approach is to rely on a human observer to inspect the registration results and reject images that have been registered with insufficient accuracy. In this paper, we present a methodology for evaluating the efficacy of the visual assessment of registration accuracy. Since the clinical requirements for level of registration accuracy are likely to be application dependent, we have evaluated the accuracy of the observer's estimate relative to six thresholds: 1-6 mm. The performance of the observers was evaluated relative to the registration solution obtained using external fiducial markers that are screwed into the patient's skull and that are visible in both MR and CT images. This fiducial marker system provides the gold standard for our study. Its accuracy is shown to be approximately 0.5 mm. Two experienced, blinded observers viewed five pairs of clinical MR and CT brain images, each of which had each been misregistered with respect to the gold standard solution. Fourteen misregistrations were assessed for each image pair with misregistration errors distributed between 0 and 10 mm with approximate uniformity. For each misregistered image pair each observer estimated the registration error (in millimeters) at each of five locations distributed around the head using each of three assessment methods. These estimated errors were compared with the errors as measured by the gold standard to determine agreement relative to each of the six thresholds, where agreement means that the two errors lie on the same side of the threshold. The effect of error in the gold standard itself is taken into account in the analysis of the assessment methods. The results were analyzed by means of the Kappa statistic, the agreement rate, and the area of receiver-operating-characteristic (ROC) curves. No assessment performed well at 1 mm, but all methods performed well at 2 mm and higher. For these five thresholds, two methods agreed with the standard at least 80% of the time and exhibited mean ROC areas greater than 0.84. One of these same methods exhibited Kappa statistics that indicated good agreement relative to chance (Kappa > 0.6) between the pooled observers and the standard for these same five thresholds. Further analysis demonstrates that the results depend strongly on the choice of the distribution of misregistration errors presented to the observers.     

Effects of coregistration of MR to CT images on MR stereotactic accuracy

Coregistration of different modality imaging serves to increase the ease and accuracy of stereotactic procedures. In many cases, magnetic resonance (MR) stereotaxis is supplanting computerized tomography (CT). The advantages of increased anatomical detail and multiplanar imaging afforded by MR, however, are offset by its potential inaccuracy as well as the more cumbersome and less available nature of its hardware. A system has been developed by one of the authors by which MR imaging can be performed separately without a stereotactic fiducial headring. Then, immediately prior to surgery, a stereotactic CT scan is obtained and software is used to coregister CT and MR images anatomically by matching cranial landmarks in the two scans. The authors examined this system in six patients as well as with the use of a lucite phantom. After initially coregistering CT and MR images, six separate anatomical (for the patients) and eight artificial (for the phantom) targets were compared. With coregistration, in comparison to CT fiducial scans, errors in each axis are less than or equal to 1 mm using the Cosman-Roberts-Wells system. In fact, the coregistered images are more accurate than MR fiducial images, in the anteroposterior (p = 0.001), lateral (p < 0.05), and vertical (p < 0.03) planes. Three-dimensional error was significantly less in the coregistered scans than the MR fiducial images (p < 0.005). The coregistration procedure therefore not only increases the case of MR stereotaxis but also increases its accuracy.     

Double oblique MR images of the shoulder: comparison with conventional images

PURPOSE: Because the scapula is not only slanted on transverse sections but also inclines on sagittal section, we now perform shoulder MR imaging using double oblique images (DOI), which are planes perpendicular or parallel to the long axis of the scapula obtained with oblique sagittal scout imaging. The purpose of this study was to evaluate the usefulness of double oblique shoulder MR imaging. MATERIALS AND METHODS: MR images of shoulders with operatively or arthroscopically proven lesions (20 cases) that had been examined on both conventional images (CI) and DOI were retrospectively reviewed. DOI were compared with CI not only in terms of diagnostic performance but also in their ability to identify the details of shoulder anatomy. All MR studies were done with a shoulder coil on a high-field (1.5T) unit. RESULTS: Although the accuracy of DOI in diagnosing shoulder disorders such as rotator cuff tear and labrum injury was not as good as that of CI, DOI were better for identifying or discriminating muscles and tendons of the rotator cuff, labralbicipital junction and anterior band of the inferior gleno-humeral ligament, and for recognizing the correct position of the glenoid labrum. CONCLUSION: MR double oblique imaging of the shoulder provides more detailed information about shoulder anatomy and disorders than conventional imaging.     

A systematic approach for interpreting MR images of the seizure patient

A systematic approach needs to be used to review MR scans in epilepsy patients to avoid the common pitfalls engendered by the subtle nature of many epileptogenic lesions. One should always evaluate the hippocampus regardless of other MR findings to avoid missing dual abnormalities. False-positive and false-negative diagnosis of hippocampal sclerosis can be avoided by evaluating the hippocampus after correcting for head rotation [by assessing the internal auditory canals and atria). Periventricular heterotopia can be successfully diagnosed by systematically studying the periventricular regions, especially those adjacent to the atria of the lateral ventricles. Gray matter lateral to the ventricles (excluding the caudate nucleus) is always an abnormal finding. Sulcal and cortical morphologic abnormalities are particularly difficult to diagnose unless a high index of suspicion is maintained. Cortical thickening is indicative of a developmental anomaly and should be screened in an organized manner. Because epilepsy is generally a cortical process, one must search for subtle cortical abnormalities, including focal atrophic abnormalities and lesions without mass effect. Diligence will offer its own rewards.     

Artefacts in MR images of electrical current distribution

Artefacts in images of electrical current distribution measured by magnetic resonance imaging (MRI) are presented. These artefacts, caused by the effects of the connecting cables used to apply the electrical current to the object during the MRI scan, can lead to an underestimation of the magnitude of the electrical current measured in the experiment. The size of this underestimate depends upon the experimental geometry, but in the experiments and simulations described here, performed with a cylindrical phantom approximately 8 cm long and 8 cm in diameter, the difference can be 27%. These artefacts are reduced if the connecting cables are rigidly fixed to the object so that the induced magnetic field is correctly measured during the MRI experiment.     

Benign and malignant musculoskeletal lesions: dynamic contrast-enhanced MR imaging--parametric "first-pass" images depict tissue vascularization and perfusion

PURPOSE: To assess the diagnostic value of parametric magnetic resonance images that display the first pass (FP) of gadopentetate dimeglumine. MATERIALS AND METHODS: Dynamic, contrast material-enhanced, fast low-angle shot MR imaging was performed of 100 musculoskeletal lesions in 36 female and 64 male patients (aged 9-81 years [mean, 34 years]). The highest slope value of the time-intensity curve was calculated pixel by pixel and displayed on the FP image, with a gray-scale value equal to the slope value. Tissue vascularization and perfusion were evaluated at pathologic and angiographic examinations in 33 lesions. RESULTS: A significant difference (P < .001) was found between the FP slope values of benign (mean, 36.2% per second) and malignant (mean, 67.4% per second) lesions. FP images depicted tissue vascularization and perfusion rather than benignity or malignancy, because there is an overlap in the slope values of highly vascular benign lesions and malignant lesions. CONCLUSION: FP images provide diagnostically useful information by depicting tissue vascularization and perfusion.     

MR image-guided portal verification for brain treatment field

PURPOSE: To investigate a method for the generation of digitally reconstructed radiographs directly from MR images (DRR-MRI) to guide a computerized portal verification procedure. METHODS AND MATERIALS: Several major steps were developed to perform an MR image-guided portal verification procedure. Initially, a wavelet-based multiresolution adaptive thresholding method was used to segment the skin slice-by-slice in MR brain axial images. Some selected anatomical structures, such as target volume and critical organs, were then manually identified and were reassigned to relatively higher intensities. Interslice information was interpolated with a directional method to achieve comparable display resolution in three dimensions. Next, a ray-tracing method was used to generate a DRR-MRI image at the planned treatment position, and the ray tracing was simply performed on summation of voxels along the ray. The skin and its relative positions were also projected to the DRR-MRI and were used to guide the search of similar features in the portal image. A Canny edge detector was used to enhance the brain contour in both portal and simulation images. The skin in the brain portal image was then extracted using a knowledge-based searching technique. Finally, a Chamfer matching technique was used to correlate features between DRR-MRI and portal image. RESULTS: The MR image-guided portal verification method was evaluated using a brain phantom case and a clinical patient case. Both DRR-CT and DRR-MRI were generated using CT and MR phantom images with the same beam orientation and then compared. The matching result indicated that the maximum deviation of internal structures was less than 1 mm. The segmented results for brain MR slice images indicated that a wavelet-based image segmentation technique provided a reasonable estimation for the brain skin. For the clinical patient case with a given portal field, the MR image-guided verification method provided an excellent match between features in both DRR-MRI and portal image. Moreover, target volume could be accurately visualized in the DRR-MRI and mapped over to the corresponding portal image for treatment verification. The accuracy of DRR-MRI was also examined by comparing it to the corresponding simulation image. The matching results indicated that the maximum deviation of anatomical features was less than 2.5 mm. CONCLUSION: A method for MR image-guided portal verification of brain treatment field was developed. Although the radiographic appearance in the DRR-MRI is different from that in the portal image, DRR-MRI provides essential anatomical features (landmarks and target volume) as well as their relative locations to be used as references for computerized portal verification.     

A multiresolution approach for contour extraction from brain images

Many image registration methods use head surface, brain surface, or inner/outer surface of the skull to estimate rotation and translation parameters. The inner surface of the skull is also used for intracranial volume segmentation which is considered the first step in segmentation and analysis of brain images. The surface is usually characterized by a set of edge or contour points extracted from cross-sectional images. Automatic extraction of contour points is complicated by discontinuity of edges in the back of the eyes and ears and sometimes by a previous surgery or an inadequate field of view. We have developed an automated method for contour extraction that connects discontinuities using a multiresolution pyramid. Steps of the method are: (1) Contour points are found by an edge-tracking algorithm; (2) A multiresolution pyramid of contour points is constructed; (3) Contour points of reduced images are found; (4) From the continuous contour found at the lowest resolution, contour points at a higher resolution are found; (5) Step 4 is repeated until contour points at the highest resolution (original image) are found. The method runs fast and has been successful in extracting contours from MRI and CT images. We illustrate the method and its performance using MRI and CT images of the human brain.     

MR of the brain in children

MR imaging has firmly established its place as the cornerstone of pediatric neuroimaging. Recent advances in MR imaging have led to decreased imaging time, high resolution studies, and new methods for obtaining tissue contrast. Magnetic resonance angiography (MRA) now obviates the need for angiography in some children, although its extended role is still to be defined. Normal and abnormal development and myelination patterns have been further defined with MR imaging. The patterns of brain injury resulting from hypoxia and ischemia vary with the degree of the insult as well as the gestational age of the child. These patterns of hypoxic-ischemic encephalopathy can be analyzed to determine when the insult occurred. Neuronal migration disorders and phakomatoses can be diagnosed with confidence at an early age, thus facilitating genetic counseling. MR imaging can detect the most common lesions associated with childhood epilepsy, such as hippocampal sclerosis, focal cortical dysplasias, and low-grade tumors. Other areas, including pediatric AIDS, toxicity-related injury, metabolic/mitochondrial conditions, and disorders associated with iatrogenic injury, can be diagnosed with MR. Spectroscopy provides information that should prove useful in evaluating and monitoring neuronal and other brain tissue disorders in children.