Manmeen Kaur

Performance of an automated segmentation algorithm for 3D MR renography

The accuracy and precision of an automated graph‐cuts (GC) segmentation technique for dynamic contrast‐enhanced (DCE) 3D MR renography (MRR) was analyzed using 18 simulated and 22 clinical datasets. For clinical data, the error was 7.2 ± 6.1 cm3 for the cortex and 6.5 ± 4.6 cm3 for the medulla. The precision of segmentation was 7.1 ± 4.2 cm3 for the cortex and 7.2 ± 2.4 cm3 for the medulla. Compartmental modeling of kidney function in 22 kidneys yielded a renal plasma flow (RPF) error of 7.5% ± 4.5% and single‐kidney GFR error of 13.5% ± 8.8%. The precision was 9.7% ± 6.4% for RPF and 14.8% ± 11.9% for GFR. It took 21 min to segment one kidney using GC, compared to 2.5 hr for manual segmentation. The accuracy and precision in RPF and GFR appear acceptable for clinical use. With expedited image processing, DCE 3D MRR has the potential to expand our knowledge of renal function in individual kidneys and to help diagnose renal insufficiency in a safe and noninvasive manner. Magn Reson Med 57:1159–1167, 2007.

Quantitative determination of Gd-DTPA concentration in T1-weighted MR renography studies.

A method for calculating contrast agent concentration from MR signal intensity (SI) was developed and validated for T1‐weighted MR renography (MRR) studies. This method is based on reference measurements of SI and relaxation time T1 in a Gd‐DTPA‐doped water phantom. The same form of SI vs. T1 dependence was observed in human tissues. Contrast concentrations calculated by the proposed method showed no bias between 0 and 1 mM, and agreed better with the reference values derived from direct T1 measurements than the concentrations calculated using the relative signal method. Phantom‐based conversion was used to determine the contrast concentrations in kidney tissues of nine patients who underwent dynamic Gd‐DTPA‐enhanced 3D MRR at 1.5T and 99mTc‐DTPA radionuclide renography (RR). The concentrations of both contrast agents were found to be close in magnitude and showed similar uptake and washout behavior. As shown by Monte Carlo simulations, errors in concentration due to SI noise were below 10% for SNR = 20, while a 10% error in precontrast T1 values resulted in a 12–17% error for concentrations between 0.1 and 1 mM. The proposed method is expected to be particularly useful for assessing regions with highly concentrated contrast. Magn Reson Med 57:1012–1018, 2007.

Pulmonary Nodules: Growth Rate Assessment in Patients by Using Serial CT and Three-dimensional Volumetry

PURPOSE To determine the precision of a three-dimensional (3D) method for measuring the growth rate of solid and subsolid nodules and its ability to detect abnormal growth rates. MATERIALS AND METHODS This study was approved by the Institutional Research Board and was HIPAA compliant. Informed consent was waived. The growth rates of 123 lung nodules in 59 patients who had undergone lung cancer screening computed tomography (CT) were measured by using a 3D semiautomated computer-assisted volume method. Clinical stability was established with long-term CT follow-up (mean, 6.4 years±1.9 [standard deviation]; range, 2.0-8.5 years). A mean of 4.1 CT examinations per patient±1.2 (range, two to seven CT examinations per patient) was analyzed during 2.4 years±0.5 after baseline CT. Nodule morphology, attenuation, and location were characterized. The analysis of standard deviation of growth rate in relation to time between scans yielded a normative model for detecting abnormal growth. RESULTS Growth rate precision increased with greater time between scans. Overall estimate for standard deviation of growth rate, on the basis of 939 growth rate determinations in clinically stable nodules, was 36.5% per year. Peripheral location (P=.01; 37.1% per year vs 25.6% per year) and adjacency to pleural surface (P=.05; 38.9% per year vs 34.0% per year) significantly increased standard deviation of growth rate. All eight malignant nodules had an abnormally high growth rate detected. By using 3D volumetry, growth rate-based diagnosis of malignancy was made at a mean of 183 days±158, compared with radiologic or clinical diagnosis at 344 days±284. CONCLUSION A normative model derived from the variability of growth rates of nodules that were stable for an average of 6.4 years may enable identification of lung cancer.

What causes diminished corticomedullary differentiation in renal insufficiency

To investigate whether the loss of corticomedullary differentiation (CMD) on T1‐weighted MR images due to renal insufficiency can be attributed to changes in T1 values of the cortex, medulla, or both.

Renal magnetic resonance imaging.

Purpose of reviewCurrent magnetic resonance imaging systems allow the visualization of normal and diseased kidney, with exquisite resolution of renal structures. Dynamic contrast magnetic resonance imaging has the potential, unique among all noninvasive modalities, to differentiate diseases that affect different portions of the vascular-nephron system. This article reviews the most important recently published studies in selected topics chosen because of their clinical relevance or potential for technical developments. Recent findingsMagnetic resonance imaging is used increasingly to evaluate renal masses, the prenatal genitourinary system, urinary obstruction and infection, renal vasculature, and the kidneys of transplant donors and recipients. Dynamic contrast magnetic resonance renography based on gadolinium chelated to diethylenetriamine pentaacetic acid, a safe (non-nephrotoxic) paramagnetic agent, emerges as the functional renal imaging modality of choice. Both perfusion and filtration rates can be assessed in individual kidney. SummaryMagnetic resonance imaging has the potential to provide a complete anatomic, physiologic, kidney-specific evaluation. With future advances in automated image analysis methods we can expect functional renal magnetic resonance imaging to play an influential role in management of renal disease.

Automatic 4-D registration in dynamic MR renography based on over-complete dyadic wavelet and fourier transforms

Dynamic contrast-enhanced 4-D MR renography has the potential for broad clinical applications, but suffers from respiratory motion that limits analysis and interpretation. Since each examination yields at least over 10 20 serial 3-D images of the abdomen, manual registration is prohibitively labor-intensive. Besides in-plane motion and translation, out-of-plane motion and rotation are observed in the image series. In this paper, a novel robust and automated technique for removing out-of-plane translation and rotation with sub-voxel accuracy in 4-D dynamic MR images is presented. The method was evaluated on simulated motion data derived directly from a clinical patients data. The method was also tested on 24 clinical patient kidney data sets. Registration results were compared with a mutual information method, in which differences between manually co-registered time-intensity curves and tested time-intensity curves were compared. Evaluation results showed that our method agreed well with these ground truth data

Automatic 4-D Registration in Dynamic MR Renography

Dynamic contrast-enhanced 4-D MR renography has the potential for broad clinical applications, but suffers from respiratory motion that limits analysis and interpretation. Since each examination yields at least over 10 - 20 serial 3-D images of the abdomen, manual registration is prohibitively labor-intensive. Besides in-plane motion and translation, out-of-plane motion and rotation are observed in the image series. In this paper, a novel robust and automated technique for removing out-of-plane translation and rotation with sub-voxel accuracy in 4-D dynamic MR images is presented. The method was evaluated on simulated motion data derived directly from a clinical patients data. The method was also tested on 24 clinical patient kidney data sets. Registration results were compared with a mutual information method, in which differences between manually co-registered time-intensity curves and tested time-intensity curves were compared. Evaluation results showed that our method agreed well with these ground truth data.

The Self-Overlap Method for Assessment of Lung Nodule Morphology in Chest CT

Surface morphology is an important indicator of malignant potential for solid-type lung nodules detected at CT, but is difficult to assess subjectively. Automated methods for morphology assessment have previously been described using a common measure of nodule shape, representative of the broad class of existing methods, termed area-to-perimeter-length ratio (APR). APR is static and thus highly susceptible to alterations by random noise and artifacts in image acquisition. We introduce and analyze the self-overlap (SO) method as a dynamic automated morphology detection scheme. SO measures the degree of change of nodule masks upon Gaussian blurring. We hypothesized that this new metric would afford equally high accuracy and superior precision than APR. Application of the two methods to a set of 119 patient lung nodules and a set of simulation nodules showed our approach to be slightly more accurate and on the order of ten times as precise, respectively. The dynamic quality of this new automated metric renders it less sensitive to image noise and artifacts than APR, and as such, SO is a potentially useful measure of cancer risk for solid-type lung nodules detected on CT.