Brain abnormalities in bipolar disorder detected by quantitative T1ρ mapping.

Abnormal metabolism has been reported in bipolar disorder, however, these studies have been limited to specific regions of the brain. To investigate whole-brain changes potentially associated with these processes, we applied a magnetic resonance imaging technique novel to psychiatric research, quantitative mapping of T1 relaxation in the rotating frame (T1ρ). This method is sensitive to proton chemical exchange, which is affected by pH, metabolite concentrations and cellular density with high spatial resolution relative to alternative techniques such as magnetic resonance spectroscopy and positron emission tomography. Study participants included 15 patients with bipolar I disorder in the euthymic state and 25 normal controls balanced for age and gender. T1ρ maps were generated and compared between the bipolar and control groups using voxel-wise and regional analyses. T1ρ values were found to be elevated in the cerebral white matter and cerebellum in the bipolar group. However, volumes of these areas were normal as measured by high-resolution T1- and T2-weighted magnetic resonance imaging. Interestingly, the cerebellar T1ρ abnormalities were normalized in participants receiving lithium treatment. These findings are consistent with metabolic or microstructural abnormalities in bipolar disorder and draw attention to roles of the cerebral white matter and cerebellum. This study highlights the potential utility of high-resolution T1ρ mapping in psychiatric research.

Image-based dose planning of intracavitary brachytherapy: registration of serial-imaging studies using deformable anatomic templates.

PURPOSE: To demonstrate that high-dimensional voxel-to-voxel transformations, derived from continuum mechanics models of the underlying pelvic tissues, can be used to register computed tomography (CT) serial examinations into a single anatomic frame of reference for cumulative dose calculations. METHODS AND MATERIALS: Three patients with locally advanced cervix cancer were treated with CT-compatible intracavitary (ICT) applicators. Each patient underwent five volumetric CT examinations: before initiating treatment, and immediately before and after the first and second ICT insertions, respectively. Each serial examination was rigidly registered to the patient's first ICT examination by aligning the bony anatomy. Detailed nonrigid alignment for organs (or targets) of interest was subsequently achieved by deforming the CT exams as a viscous-fluid, described by the Navier-Stokes equation, until the coincidence with the corresponding targets on CT image was maximized. In cases where ICT insertion induced very large and topologically complex rearrangements of pelvic organs, e.g., extreme uterine canal reorientation following tandem insertion, a viscous-fluid-landmark transformation was used to produce an initial registration. RESULTS: For all three patients, reasonable registrations for organs (or targets) of interest were achieved. Fluid-landmark initialization was required in 4 of the 11 registrations. Relative to the best rigid bony landmark alignment, the viscous-fluid registration resulted in average soft-tissue displacements from 2.8 to 28.1 mm, and improved organ coincidence from the range of 5.2% to 72.2% to the range of 90.6% to 100%. Compared to the viscous-fluid transformation, global registration of bony anatomy mismatched 5% or more of the contoured organ volumes by 15-25 mm. CONCLUSION: Pelvic soft-tissue structures undergo large deformations and displacements during the external-beam and multiple-ICT course of radiation therapy for locally advanced cervix cancer. These changes cannot be modeled by the conventional rigid landmark transformation method. In the current study, we found that the deformable anatomic template registration method, based on continuum-mechanics models of deformation, successfully described these large anatomic shape changes before and after ICT. These promising modeling results indicate that realistic registration of the cumulative dose distribution to the organs (or targets) of interest for radiation therapy of cervical cancers is achievable.

Consistent image registration.

This paper presents a new method for image registration based on jointly estimating the forward and reverse transformations between two images while constraining these transforms to be inverses of one another. This approach produces a consistent set of transformations that have less pairwise registration error, i.e., better correspondence, than traditional methods that estimate the forward and reverse transformations independently. The transformations are estimated iteratively and are restricted to preserve topology by constraining them to obey the laws of continuum mechanics. The transformations are parameterized by a Fourier series to diagonalize the covariance structure imposed by the continuum mechanics constraints and to provide a computationally efficient numerical implementation. Results using a linear elastic material constraint are presented using both magnetic resonance and X-ray computed tomography image data. The results show that the joint estimation of a consistent set of forward and reverse transformations constrained by linear-elasticity give better registration results than using either constraint alone or none at all.

Consistent landmark and intensity-based image registration.

Two new consistent image registration algorithms are presented: one is based on matching corresponding landmarks and the other is based on matching both landmark and intensity information. The consistent landmark and intensity registration algorithm produces good correspondences between images near landmark locations by matching corresponding landmarks and away from landmark locations by matching the image intensities. In contrast to similar unidirectional algorithms, these new consistent algorithms jointly estimate the forward and reverse transformation between two images while minimizing the inverse consistency error-the error between the forward (reverse) transformation and the inverse of the the reverse (forward) transformation. This reduces the ambiguous correspondence between the forward and reverse transformations associated with large inverse consistency errors. In both algorithms a thin-plate spline (TPS) model is used to regularize the estimated transformations. Two-dimensional (2-D) examples are presented that show the inverse consistency error produced by the traditional unidirectional landmark TPS algorithm can be relatively large and that this error is minimized using the consistent landmark algorithm. Results using 2-D magnetic resonance imaging data are presented that demonstrate that using landmark and intensity information together produce better correspondence between medical images than using either landmarks or intensity information alone.

Volumetric transformation of brain anatomy.

This paper presents diffeomorphic transformations of three-dimensional (3-D) anatomical image data of the macaque occipital lobe and whole brain cryosection imagery and of deep brain structures in human brains as imaged via magnetic resonance imagery. These transformations are generated in a hierarchical manner, accommodating both global and local anatomical detail. The initial low-dimensional registration is accomplished by constraining the transformation to be in a low-dimensional basis. The basis is defined by the Green's function of the elasticity operator placed at predefined locations in the anatomy and the eigenfunctions of the elasticity operator. The high-dimensional large deformations are vector fields generated via the mismatch between the template and target-image volumes constrained to be the solution of a Navier-Stokes fluid model. As part of this procedure, the Jacobian of the transformation is tracked, insuring the generation of diffeomorphisms. It is shown that transformations constrained by quadratic regularization methods such as the Laplacian, biharmonic, and linear elasticity models, do not ensure that the transformation maintains topology and, therefore, must only be used for coarse global registration.

Retrospective evaluation of intersubject brain registration.

Although numerous methods to register brains of different individuals have been proposed, no work has been done, as far as we know, to evaluate and objectively compare the performances of different nonrigid (or elastic) registration methods on the same database of subjects. In this paper, we propose an evaluation framework, based on global and local measures of the relevance of the registration. We have chosen to focus more particularly on the matching of cortical areas, since intersubject registration methods are dedicated to anatomical and functional normalization, and also because other groups have shown the relevance of such registration methods for deep brain structures. Experiments were conducted using 6 methods on a database of 18 subjects. The global measures used show that the quality of the registration is directly related to the transformation's degrees of freedom. More surprisingly, local measures based on the matching of cortical sulci did not show significant differences between rigid and non rigid methods.

3D brain mapping using a deformable neuroanatomy.

This paper presents two different mathematical methods that can be used separately or in conjunction to accommodate shape variabilities between normal human neuroanatomies. Both methods use a digitized textbook to represent the complex structure of a typical normal neuroanatomy. Probabilistic transformations on the textbook coordinate system are defined to accommodate shape differences between the textbook and images of other normal neuroanatomies. The transformations are constrained to be consistent with the physical properties of deformable elastic solids in the first method and those of viscous fluids in the second. Results presented in this paper demonstrate how a single deformable textbook can be used to accommodate normal shape variability.

Stereological MRI volumetry of the frontal lobe.

Stereology was used to measure frontal lobe volume on magnetic resonance imaging (MRI) scans in a multi-observer repeated-measures trial in 17 adults. Prior to measurement, MR image volumes were reoriented into coronal sections perpendicular to the bicommissural plane. Three observers blinded to subject identify repeatedly used fixed grid stereology to estimate frontal lobe volumes, defined as all sections of the frontal lobe anterior to the anterior commissure. The lateral ventricles were excluded. Stereological measurement yielded high repeatability and precision, and was time efficient for the raters. The coefficient of error was 0.03. The inter-rater correlation coefficient = 0.95 for three raters; intra-rater correlation coefficients = 0.95-0.98. A comparison was made between stereological and traditional edge tracing measurement of the frontal lobe volumes. The overall correlation between the two methods was 0.95. The use of internal landmarks to define orientation and 3-D orthogonal views to define frontal lobe boundaries on 3-D images was critical to obtaining repeatable measurements. Frontal lobe volumetry by brain MR used to estimate small differences postulated to occur in certain psychiatric and neurologic disorders requires high precision and repeatability. Stereology, a semi-automated method, can reliably estimate frontal lobe volumes. This method may distinguish small frontal lobe volume differences within individuals and between groups.

Deformable templates using large deformation kinematics.

A general automatic approach is presented for accommodating local shape variation when mapping a two-dimensional (2-D) or three-dimensional (3-D) template image into alignment with a topologically similar target image. Local shape variability is accommodated by applying a vector-field transformation to the underlying material coordinate system of the template while constraining the transformation to be smooth (globally positive definite Jacobian). Smoothness is guaranteed without specifically penalizing large-magnitude deformations of small subvolumes by constraining the transformation on the basis of a Stokesian limit of the fluid-dynamical Navier-Stokes equations. This differs fundamentally from quadratic penalty methods, such as those based on linearized elasticity or thin-plate splines, in that stress restraining the motion relaxes over time allowing large-magnitude deformations. Kinematic nonlinearities are inherently necessary to maintain continuity of structures during large-magnitude deformations, and are included in all results. After initial global registration, final mappings are obtained by numerically solving a set of nonlinear partial differential equations associated with the constrained optimization problem. Automatic regridding is performed by propagating templates as the nonlinear transformations evaluated on a finite lattice become singular. Application of the method to intersubject registration of neuroanatomical structures illustrates the ability to account for local anatomical variability.

Relationship between bone and muscles of mastication in hemifacial microsomia.

The relationship between the bone and muscles of mastication in hemifacial microsomia was studied using three-dimensional volumetric computed tomography scans and image processing techniques. High resolution head computed tomography scans were obtained from 31 patients with unilateral hemifacial microsomia and eight normal patients. Using three-dimensional volume renderings of bone, mandibular deformities in patients with hemifacial microsomia were classified using the Pruzansky system. For each patient, specific craniofacial bones (temporal bone, maxilla mandible) and the muscles of mastication (masseter, temporalis and lateral and medial pterygoid) were segmented bilaterally from the image volume for independent display and volume measurement. Volumes were expressed as the ratio of the affected: unaffected sides. For the masseter and temporalis, the relationship between muscular hypoplasia and osseous hypoplasia in its origin and insertion was studied by plotting affected:unaffected bone volume as a function of affected:unaffected muscle volume for each muscle, bone of origin, bone of insertion triplet. The volumes of the pterygoid muscles were compared with hemimandibular volumes. The precision of object segmentations was examined by repetitive definition tasks, whereas the accuracy of volume measurement was tested by scanning custom-made phantom objects and comparing digital to physical object volume measurements. Volume measurements performed using these techniques were both accurate and precise. In hemifacial microsomia, the extent of hypoplasia of specific muscles of mastication predicted the extent of dysplasia in their osseous origin and insertion. However, the reverse was not true. The extent of hypoplasia of the facial bones did not necessarily predict the extent of hypoplasia in the attached muscles of mastication. Pruzansky grade of the mandible described the degree of mandibular hypoplasia on the affected side, but was inconsistent in its prediction of volume decrease of the other facial bones.

The craniofacial anomalies archive at St. Louis Children's Hospital: 20 years of craniofacial imaging experience.

This article describes how the Craniofacial Imaging Laboratory at the Cleft Palate and Craniofacial Deformities Institute, St. Louis Children's Hospital, Washington University Medical Center, has developed an electronic archive for the storage of computed tomography image digital data that is independent of scanner hardware and independent of units of storage media (i.e., floppy disks and optical disks). The archive represents one of the largest repositories of high-quality computed tomography data of children with craniofacial deformities in the world. Archiving reconstructed image data is essential for comparative imaging, surgical simulation, quantitative analysis, and use with solid model fabrication (e.g., stereolithography). One tertiary craniofacial center's experience in the establishment and maintenance of such an archive through three generations of storage technology is reported. The current archive is housed on an external 35-GB hard drive attached to a Windows-based desktop server. Data in the archive were categorized by specific demographics into groups of patients, number of scans, and diagnoses. The Craniofacial Imaging Laboratory archive currently contains computed tomography image digital data for 1827 individual scans. The earliest scan was done in 1980; the most recently stored scan for the purposes of this report occurred in May of 2000. The average number of scans archived per complete year was 94, with a range of 59 to 138. Of the 1827 total scans, 74 percent could be classified into specific diagnostic categories. The majority of the archive (55 percent) is composed of the following five diagnoses: sagittal synostosis (17 percent), unilateral coronal synostosis (11 percent), hemifacial microsomia (10 percent), plagiocephaly without synostosis (10 percent), and metopic synostosis (7 percent). Storage of computed tomography image data in a digital archive currently allows for continuous upgrading of image display and analysis and facilitates longitudinal and cross-sectional studies, both intramural and extramural. Internet access for clinical and research purposes is feasible, but contingent on protection of patient confidentiality. The future of digital imaging regarding craniofacial computed tomography scan storage and processing is also discussed.

Surgical imaging systems.

Imaging i.n.surgery. is used for diagnosis, planning, intraoperative navigation and post-operative evaluation, DIgItal medical imaging modalities mclude computed tomography (CT), magnetic resonance Imaging (MRI), MR therapy (MRT), fluoroscopy and ultrasound.t These modalities are applied singly or jointly (multimodality) Surgical requirements differ according to the nature of intervention, and real-time guidance? is sometimes needed such that a sequence of images is generated and displayed as acquired, Soft copy display on CRTscreens is satisfactory for intraoperative use, while hardcopy film images or physical replica modeling may be needed in other cases. Computed tomography/ developed more than 20 years ago, remains important in craniofacial'v" and orthopedic surgery." Newer imaging systems, especially ultrasound, magnetic resonance imaging,'6 and digital fluoroscopy are used for neurosurgery, oncology, cardiothoracic and abdominal surgery. Each modality offers specific qualities that subserve specific needs in diagnosis, planning, intraoperative navigation and evaluation (Table 1).

WE-E-213CD-05: A Non-Rigid Image Registration Algorithm That Accommodates Organ Segmentation Error.

PURPOSE: To introduce a new deformable image registration algorithm based on surface matching that accommodates organ delineation error in daily Cone-beam CT images based on a priori knowledge of inter-observer segmentation uncertainty. METHODS: The dataset includes four prostate cancer patients who underwent primary external beam radiotherapy and had tumors that were confined to the prostate. All imaging was performed without intravenous contrast. Organ surface segmentation errors in a multiple observer-contouring study on the pelvic organs in Fan-beam CT (FBCT) and Cone-beam CT (CBCT) were estimated from the training dataset. A novel deformable image registration algorithm is presented where the organ surface matching is penalized by this error. Portions of the organ surface that are delineated reliably are used to guide the registration whereas the portions that are highly uncertain are ignored. This approach reduces the impact of delineation errors in CBCT. An evaluation experiment compares three algorithms, namely intensity-only registration (INT), equally-weighted surface and image registration (EWSIR) and the proposed uncertainty- weighted surface and image registration. RESULTS: The surface dissimilarity was reduced from 0.172 to 0.134, 0.043 and 0.044 respectively after registration. The Jacobian of the transformation found by the proposed method was closer to one than that of EWSIR in the prostate. CONCLUSIONS: In prostate external-beam radiotherapy, slice-by-slice 2D manual contouring has variable spatial accuracy. For deformable image registration methods that match segmented surfaces, regions of high inaccuracy can misguide the registration. In contrast to the image registration methods where the FBCT and CBCT surfaces (or other features) are assumed to be exact, our method takes this uncertainty into account. Preliminary results show an improved registration performance suggesting a potential use in IGRT. This work was supported by National Cancer Institute Grant No. P01 CA 116602.

Mathematical textbook of deformable neuroanatomies.

Mathematical techniques are presented for the transformation of digital anatomical textbooks from the ideal to the individual, allowing for the representation of the variabilities manifest in normal human anatomies. The ideal textbook is constructed on a fixed coordinate system to contain all of the information currently available about the physical properties of neuroanatomies. This information is obtained via sensor probes such as magnetic resonance, as well as computed axial and emission tomography, along with symbolic information such as white- and gray-matter tracts, nuclei, etc. Human variability associated with individuals is accommodated by defining probabilistic transformations on the textbook coordinate system, the transformations forming mathematical translation groups of high dimension. The ideal is applied to the individual patient by finding the transformation which is consistent with physical properties of deformable elastic solids and which brings the coordinate system of the textbook to that of the patient. Registration, segmentation, and fusion all result automatically because the textbook carries symbolic values as well as multisensor features.

Hippocampal MR imaging morphometry by means of general pattern matching.

PURPOSE: To determine the repeatability and validity of a pattern-matching method for the segmentation and measurement of hippocampi on magnetic resonance (MR) images. MATERIALS AND METHODS: Comparable two-dimensional MR images obtained in 18 subjects (nine healthy control subjects [six men, three women; aged 24-54 years] and nine patients with schizophrenia [six men, three women; aged 22-61 years]) were twice segmented manually and twice segmented by using pattern matching with digital atlas transformation. The atlas transformation was accomplished in two steps: global followed by local matching. Global matching was performed with use of landmarks; local matching was performed with use of a viscous fluid model. RESULTS: The mean percentage of difference between two atlas-based measurements was 1.33% +/- 1.23 (+/- standard deviation); that between two manual measurements was 4.67% +/- 4.71. The validity of the atlas transformation measurements was demonstrated by means of the high correlation (intraclass correlation coefficient = .96) with manual segmentation measurements. Schizophrenic hippocampal areas tended to be smaller; however, no differences in hippocampal shape were found between patients with schizophrenia and patients with control subjects. CONCLUSION: General pattern matching of a digital brain atlas to an individual MR image is a mathematically robust method of measurement that is reproducible and less variable than manual measurement.

Three-dimensional hippocampal MR morphometry with high-dimensional transformation of a neuroanatomic atlas.

PURPOSE: To test automated three-dimensional magnetic resonance (MR) imaging morphometry of the human hippocampus, to determine the potential gain in precision compared with conventional manual morphometry. MATERIAL AND METHODS: A canonical three-dimensional MR image atlas was used as a deformable template and automatically matched to three-dimensional MR images of 10 individuals (five healthy and five schizophrenic subjects). A subvolume containing the hippocampus was defined by using 16 landmarks that constrained the automated search for hippocampal boundaries. Transformation of the hippocampus template was automatically performed by using global pattern matching through a sequence of low-then high-dimensional translations, rotations, and scalings. RESULTS: The average test-retest volume difference measured with the automatic method was 3.1%, compared with the manual test-retest difference of 7.1%. Correlation between automated and manually determined volumes demonstrated the validity of the automated technique (intraclass correlation coefficient = .86). CONCLUSION: The automated method estimates hippocampal volumes with less variability (ie, lower variance) than that of manual out-lining.