Usha Sinha

Efficient Multilevel Brain Tumor Segmentation With Integrated Bayesian Model Classification

We present a new method for automatic segmentation of heterogeneous image data that takes a step toward bridging the gap between bottom-up affinity-based segmentation methods and top-down generative model based approaches. The main contribution of the paper is a Bayesian formulation for incorporating soft model assignments into the calculation of affinities, which are conventionally model free. We integrate the resulting model-aware affinities into the multilevel segmentation by weighted aggregation algorithm, and apply the technique to the task of detecting and segmenting brain tumor and edema in multichannel magnetic resonance (MR) volumes. The computationally efficient method runs orders of magnitude faster than current state-of-the-art techniques giving comparable or improved results. Our quantitative results indicate the benefit of incorporating model-aware affinities into the segmentation process for the difficult case of glioblastoma multiforme brain tumor.

In vivo diffusion-weighted MRI of the breast: Potential for lesion characterization

To investigate the potential of apparent diffusion coefficients (ADCs) in characterizing breast lesions in vivo.

Relationships Between Choline Magnetic Resonance Spectroscopy, Apparent Diffusion Coefficient and Quantitative Histopathology in Human Glioma

This study sought to correlate quantitative presurgical proton magnetic resonance spectroscopic imaging (1H- MRSI) and diffusion imaging (DI) results with quantitative histopathological features of resected glioma tissue. The primary hypotheses were (1) glioma choline signal correlates with cell density, (2) glioma apparent diffusion coefficient (ADC) correlates inversely with cell density, (3) glioma choline signal correlates with cell proliferative index. Eighteen adult glioma patients were preoperatively imaged with 1H-MRSI and DI as part of clinically-indicated MRI evaluations. Cell density and proliferative index readings were made on surgical specimens obtained at surgery performed within 12 days of the radiologic scans. The resected tissue location was identified by comparing preoperative and postoperative MRI. The tumor to contralateral normalized choline signal ratio (nCho) and the ADC from resected tumor regions were measured from the preoperative imaging data. Counts of nuclei per high power field in 5–10 fields provided a quantitative measure of cell density. MIB-1 immunohistochemistry provided an index of the proportion of proliferating cells. There was a statistically significant inverse linear correlation between glioma ADC and cell density. There was also a statistically significant linear correlation between the glioma nCho and the cell density. The nCho measure did not significantly correlate with proliferative index. The results indicate that both ADC and spectroscopic choline measures are related to glioma cell density. Therefore they may prove useful for differentiating dense cellular neoplastic lesions from those that contain large proportions of acellular necrotic space.

In vivo diffusion tensor imaging of the human calf muscle

To demonstrate the feasibility of in vivo calf muscle fiber tracking in human subjects.

Inverse correlation between choline magnetic resonance spectroscopy signal intensity and the apparent diffusion coefficient in human glioma

Magnetic resonance spectroscopy and diffusion magnetic resonance imaging (MRI) characteristics of human intracranial glioma were studied. Present knowledge suggests a hypothetical inverse relationship between the characteristic choline signal intensity elevation and the apparent diffusion coefficient (ADC) in glioma. Twenty cases of glioma were examined with diffusion‐weighted echoplanar imaging and proton magnetic resonance spectroscopic imaging (1H‐MRSI). A statistically significant inverse correlation between the choline signal intensity and the ADC was found (P = 0.0004) in radiologically defined tumor‐containing regions. This study is the first in which diffusion MRI and 1H‐MRSI were used to evaluate human intracranial glioma jointly. It provides insight into how to interpret choline signal intensity elevation in terms of tumor cellularity and proliferative potential when ADC images are also available. Magn Reson Med 41:2‐7, 1999.

Effect of field strength on susceptibility artifacts in magnetic resonance imaging

In magnetic resonance imaging susceptibility artifacts occur at the interface of substances with large magnetic susceptibility differences, resulting in geometric distortions of the image at those boundaries. The susceptibility artifacts are often subtle on clinical images and if not carefully examined they may lead to misdiagnosis. Magnetic susceptibility artifacts are prevalent on the boundary of air-containing paranasal sinuses, as well as bone-soft tissue interfaces in the spinal canal. The appearance of these artifacts on images from three different magnetic field strength instruments, 0.3, 0.5, and 1.5 Tesla were studied. T1- and T2-weighted spin echo and gradient recalled echo pulse sequences were selected to image a water phantom containing substances of varying susceptibilities. The effects were also studied in MR images of the head in a normal human volunteer. At any given field strength the artifacts were more prominent in the gradient echo imaging than in the corresponding spin echo pulse sequence. As expected, the distortions were also greater at higher field strengths. The results in human subjects paralleled the findings in the phantom study.

In vivo diffusion tensor imaging of the human prostate

This study demonstrates the feasibility of in vivo prostate diffusion tensor imaging (DTI) in human subjects. We implemented an EPI‐based diffusion‐weighted (DW) sequence with seven‐direction diffusion gradient sensitization, and acquired DT images from six subjects using cardiac gating with a phased‐array prostate surface coil operating in a linear mode. We calculated two indices to quantify diffusion anisotropy. The direction of the eigenvector corresponding to the leading eigenvalue was displayed by means of a color‐coding scheme. The average diffusion values of the prostate peripheral zone (PZ) and central gland (CG) were 1.95 ± 0.08 × 10–3 mm2 s and 1.53 ± 0.34 × 10–3 mm2 s, respectively. The average fractional anisotropy (FA) values for the PZ and CG were 0.46 ± 0.04 and 0.40 ± 0.08, respectively. The diffusion ellipsoid in prostate tissue was anisotropic and approximated a prolate model, as shown in the color maps of the anisotropy. Consistent with the tissue architecture, the prostate fiber orientations were predominantly in the superior–inferior (SI) direction for both the PZ and CG. This study shows the feasibility of in vivo DTI and establishes normative DT values for six subjects. Magn Reson Med 52:530–537, 2004.

Geometric distortion correction of high-resolution 3 T diffusion tensor brain images

Diffusion‐weighted images based on echo planar sequences suffer from distortions due to field inhomogeneities from susceptibility differences as well as from eddy currents arising from diffusion gradients. In this paper, a novel approach using nonlinear warping based on optic flow to correct distortions of baseline and diffusion weighted echo planar images (EPI) acquired at 3 T is presented. The distortion correction was estimated by warping the echo planar images to the anatomically correct T2‐weighted fast spin echo images (T2‐FSE). A global histogram intensity matching of the T2‐FSE precedes the base line EPI image distortion correction. A local intensity‐matching algorithm was used to transform labeled T2‐FSE regions to match intensities of diffusion‐weighted EPI images prior to distortion correction of these images. Evaluation was performed using three methods: (i) visual comparison of overlaid contours, (ii) a global mutual information index, and (iii) a local distance measure between homologous points. Visual assessment and the global index demonstrated a decrease in geometrical distortion and the distance measure showed that distortions are reduced to a subvoxel level. In conclusion, the warping algorithm is effective in reducing geometric distortions, enabling generation of anatomically correct diffusion tensor images at 3 T. Magn Reson Med, 2005.

In vivo diffusion tensor imaging of human calf muscle.

To investigate a tetrahedral diffusion gradient encoding scheme to measure the diffusion tensor in vivo for human calf muscle.

Functional Magnetic Resonance of Human Breast Tumors

Abstract: This review is focused on two relatively new developments in magnetic resonance imaging (MRI) and their application to breast lesion characterization: diffusion and perfusion MRI. Diffusion MRI measures the mobility of the water protons and thus provides a window to tissue microstructure. Perfusion MRI measures the rate at which blood is delivered to tissue and thus provides information about microvasculature. Because both tissue structure and vasculature are likely to change in disease states, measurement of diffusion and perfusion may have direct physiologic relevance. This review covers topics related to the imaging sequences, image analysis, and clinical studies for diffusion and perfusion breast MRI. Preliminary studies show that the apparent diffusion coefficient (ADC) is a marker of cell density and can distinguish malignant from benign lesions. Perfusion MR also shows promise for breast tumor characterization: malignant tumors have consistently higher relative tissue blood volumes (rTBV) than normal and benign tumors. Additional research is required with large patient cohorts to establish these two techniques on a clinical footing.