Data-driven modeling of noise time series with convolutional generative adversarial networks

Random noise arising from physical processes is an inherent characteristic of measurements and a limiting factor for most signal processing and data analysis tasks. Given the recent interest in generative adversarial networks (GANs) for data-driven modeling, it is important to determine to what extent GANs can faithfully reproduce noise in target data sets. In this paper, we present an empirical investigation that aims to shed light on this issue for time series. Namely, we assess two general-purpose GANs for time series that are based on the popular deep convolutional GAN architecture, a direct time-series model and an image-based model that uses a short-time Fourier transform data representation. The GAN models are trained and quantitatively evaluated using distributions of simulated noise time series with known ground-truth parameters. Target time series distributions include a broad range of noise types commonly encountered in physical measurements, electronics, and communication systems: band-limited thermal noise, power law noise, shot noise, and impulsive noise. We find that GANs are capable of learning many noise types, although they predictably struggle when the GAN architecture is not well suited to some aspects of the noise, e.g. impulsive time-series with extreme outliers. Our findings provide insights into the capabilities and potential limitations of current approaches to time-series GANs and highlight areas for further research. In addition, our battery of tests provides a useful benchmark to aid the development of deep generative models for time series.

Blind Measurement of Receiver System Noise

Tightly packaged receivers pose a challenge for noise measurements. Their only outputs are often diagnostic or benchmark information-"user data" that result from unknown processing. These include data rate test results, signal-to-noise ratio estimated by the receiver, and so on. Some of these are important gauges of communication viability that may be enshrined in performance and conformance specifications. Engineers can estimate these parameters based on standards and simplified system models, but there are few means to validate against physical measurements. We propose here a set of measurement techniques to complement and support models of system noise. The approach is founded on a semiparametric model of the noise response of a full-stack receiver. We probe this response experimentally by systematically perturbing signals and excess noise levels at the receiver input. The resulting technique is blind to protocol and implementation details. We introduce the design and implementation of some novel test capabilities required for these tests: a precision programmable excess noise source and a highly directive programmable attenuator. We also introduce a regression procedure to estimate system noise (or NF) from the controlled input conditions and summary statistics of the user data output. We also estimate uncertainty in the measurement by combining traditional methods with a Monte Carlo method that propagates random errors through the regression. Case studies demonstrate the measurement with consumer wireless networking and geolocation equipment. These include verification by repeatability testing and cross-comparison against Y-factor measurements.

Deep Learning Classification of 3.5 GHz Band Spectrograms with Applications to Spectrum Sensing

In the United States, the Federal Communications Commission has adopted rules permitting commercial wireless networks to share spectrum with federal incumbents in the 3.5 GHz Citizens Broadband Radio Service band. These rules require commercial systems to vacate the band when sensors detect radars operated by the U.S. military; a key example being the SPN-43 air traffic control radar. Such sensors require highly-accurate detection algorithms to meet their operating requirements. In this paper, using a library of over 14,000 3.5 GHz band spectrograms collected by a recent measurement campaign, we investigate the performance of thirteen methods for SPN-43 radar detection. Namely, we compare classical methods from signal detection theory and machine learning to several deep learning architectures. We demonstrate that machine learning algorithms appreciably outperform classical signal detection methods. Specifically, we find that a three-layer convolutional neural network offers a superior tradeoff between accuracy and computational complexity. Last, we apply this three-layer network to generate descriptive statistics for the full 3.5 GHz spectrogram library. Our findings highlight potential weaknesses of classical methods and strengths of modern machine learning algorithms for radar detection in the 3.5 GHz band.

Internal radionuclide therapy

Aim: Accurate dosimetry must be performed for each patient before therapy with unsealed radionuclides. Recently, the software tool ULMDOS was developed to facilitate planar dosimetric calculations and to support traceability and documentation as a prerequisite for good clinical practice. Here, the extended version of ULMDOS for processing of tomographic data is presented. Methods: ULMDOS is developed in IDL 6.1 (Interactive Data Language) under Windows XP/2000. Serial tomographic data can be loaded in an ECAT7 or DICOM format, and presented as maximum intensity projection. The definition of volumes of interest is supported by various tools (e.g., freehand, isocontour, polygon), region growing, and cluster analysis. Residence times are calculated from fits of the time activity data to exponential functions. Results, discussion: Quantitative 3-dimensional data allow performing a more individualized dosimetry, as problems due to organ overlay, insufficient attenuation and scatter correction in the planar approach can be avoided. For traceability, documentation, retrospective examination and later processing all data can be saved in binary or ASCII format. Dosimetric calculations can be conducted within a single environment, thus it spares the time-consuming transfer of data between different software tools.

Optimal Joint Detection and Estimation That Maximizes ROC-Type Curves

Combined detection-estimation tasks are frequently encountered in medical imaging. Optimal methods for joint detection and estimation are of interest because they provide upper bounds on observer performance, and can potentially be utilized for imaging system optimization, evaluation of observer efficiency, and development of image formation algorithms. We present a unified Bayesian framework for decision rules that maximize receiver operating characteristic (ROC)-type summary curves, including ROC, localization ROC (LROC), estimation ROC (EROC), free-response ROC (FROC), alternative free-response ROC (AFROC), and exponentially-transformed FROC (EFROC) curves, succinctly summarizing previous results. The approach relies on an interpretation of ROC-type summary curves as plots of an expected utility versus an expected disutility (or penalty) for signal-present decisions. We propose a general utility structure that is flexible enough to encompass many ROC variants and yet sufficiently constrained to allow derivation of a linear expected utility equation that is similar to that for simple binary detection. We illustrate our theory with an example comparing decision strategies for joint detection-estimation of a known signal with unknown amplitude. In addition, building on insights from our utility framework, we propose new ROC-type summary curves and associated optimal decision rules for joint detection-estimation tasks with an unknown, potentially-multiple, number of signals in each observation.

MRMC analysis of agreement studies

The purpose of this work is to present and evaluate methods based on U-statistics to compare intra- or inter-reader agreement across different imaging modalities. We apply these methods to multi-reader multi-case (MRMC) studies. We measure reader-averaged agreement and estimate its variance accounting for the variability from readers and cases (an MRMC analysis). In our application, pathologists (readers) evaluate patient tissue mounted on glass slides (cases) in two ways. They evaluate the slides on a microscope (reference modality) and they evaluate digital scans of the slides on a computer display (new modality). In the current work, we consider concordance as the agreement measure, but many of the concepts outlined here apply to other agreement measures. Concordance is the probability that two readers rank two cases in the same order. Concordance can be estimated with a U-statistic and thus it has some nice properties: it is unbiased, asymptotically normal, and its variance is given by an explicit formula. Another property of a U-statistic is that it is symmetric in its inputs; it doesn't matter which reader is listed first or which case is listed first, the result is the same. Using this property and a few tricks while building the U-statistic kernel for concordance, we get a mathematically tractable problem and efficient software. Simulations show that our variance and covariance estimates are unbiased.

A Utility/Cost Analysis of Breast Cancer Risk Prediction Algorithms

Breast cancer risk prediction algorithms are used to identify subpopulations that are at increased risk for developing breast cancer. They can be based on many different sources of data such as demographics, relatives with cancer, gene expression, and various phenotypic features such as breast density. Women who are identified as high risk may undergo a more extensive (and expensive) screening process that includes MRI or ultrasound imaging in addition to the standard full-field digital mammography (FFDM) exam. Given that there are many ways that risk prediction may be accomplished, it is of interest to evaluate them in terms of expected cost, which includes the costs of diagnostic outcomes. In this work we perform an expected-cost analysis of risk prediction algorithms that is based on a published model that includes the costs associated with diagnostic outcomes (true-positive, false-positive, etc.). We assume the existence of a standard screening method and an enhanced screening method with higher scan cost, higher sensitivity, and lower specificity. We then assess expected cost of using a risk prediction algorithm to determine who gets the enhanced screening method under the strong assumption that risk and diagnostic performance are independent. We find that if risk prediction leads to a high enough positive predictive value, it will be cost-effective regardless of the size of the subpopulation. Furthermore, in terms of the hit-rate and false-alarm rate of the of the risk-prediction algorithm, iso-cost contours are lines with slope determined by properties of the available diagnostic systems for screening.

Individualized multimodal treatment strategy for anaplastic thyroid carcinoma-Case report of long-term remission and review of literature

INTRODUCTION

The prognosis of anaplastic thyroid cancer (ATC) is poor with a mean survival time of six months following diagnosis. Despite various attempts to modify common treatment modalities including surgery, external beam radiation and chemotherapy, an effective treatment is not available yet. We report, here, a patient who achieved long-term survival based on multimodal treatment, including in vitro evaluation of drug response of his tumor cells.

PRESENTATION OF CASE

A 42 years old male patient underwent total thyroidectomy with central and lateral neck dissection for ATC (pT4b, pN0 (0/36), L0, V0, Pn1, R0 cM0 - UICC-Stage: IV b). From the tumor tissue a primary cell culture was established. While the patient received a combined radio-chemotherapy cell viability assays were performed using Sorafenib, Vandetanib und MLN8054 (Aurora kinase inhibitor) as inhibitors. Cell viability was determined by MTT-assay after 72 and 144h of treatment.

DISCUSSION

All the three compounds affected cell viability in a time- and dose dependent manner. These effects were most pronounced by Sorafenib. Based on in vitro findings, the patient was treated daily with 400mg Sorafenib for 75days. 43 months after initial diagnosis, the patient had no evidence of disease as shown by MRI, CT and FDG-PET-CT imaging.

CONCLUSION

In the setting of multimodal treatment, in vitro drug evaluation of individual tumor cells of patients might be a promising tool to ameliorate the fatal prognosis of selected ATC patients.

Seamless Insertion of Pulmonary Nodules in Chest CT Images

The availability of large medical image datasets is critical in many applications, such as training and testing of computer-aided diagnosis systems, evaluation of segmentation algorithms, and conducting perceptual studies. However, collection of data and establishment of ground truth for medical images are both costly and difficult. To address this problem, we are developing an image blending tool that allows users to modify or supplement existing datasets by seamlessly inserting a lesion extracted from a source image into a target image. In this study, we focus on the application of this tool to pulmonary nodules in chest CT exams. We minimize the impact of user skill on the perceived quality of the composite image by limiting user involvement to two simple steps: the user first draws a casual boundary around a nodule in the source, and, then, selects the center of desired insertion area in the target. We demonstrate the performance of our system on clinical samples, and report the results of a reader study evaluating the realism of inserted nodules compared to clinical nodules. We further evaluate our image blending techniques using phantoms simulated under different noise levels and reconstruction filters. Specifically, we compute the area under the ROC curve of the Hotelling observer (HO) and noise power spectrum of regions of interest enclosing native and inserted nodules, and compare the detectability, noise texture, and noise magnitude of inserted and native nodules. Our results indicate the viability of our approach for insertion of pulmonary nodules in clinical CT images.

Exact confidence intervals for channelized Hotelling observer performance in image quality studies

Task-based assessments of image quality constitute a rigorous, principled approach to the evaluation of imaging system performance. To conduct such assessments, it has been recognized that mathematical model observers are very useful, particularly for purposes of imaging system development and optimization. One type of model observer that has been widely applied in the medical imaging community is the channelized Hotelling observer (CHO), which is well-suited to known-location discrimination tasks. In the present work, we address the need for reliable confidence interval estimators of CHO performance. Specifically, we show that the bias associated with point estimates of CHO performance can be overcome by using confidence intervals proposed by Reiser for the Mahalanobis distance. In addition, we find that these intervals are well-defined with theoretically-exact coverage probabilities, which is a new result not proved by Reiser. The confidence intervals are tested with Monte Carlo simulation and demonstrated with two examples comparing X-ray CT reconstruction strategies. Moreover, commonly-used training/testing approaches are discussed and compared to the exact confidence intervals. MATLAB software implementing the estimators described in this work is publicly available at http://code.google.com/p/iqmodelo/.

Multireader multicase reader studies with binary agreement data: simulation, analysis, validation, and sizing

We treat multireader multicase (MRMC) reader studies for which a reader's diagnostic assessment is converted to binary agreement (1: agree with the truth state, 0: disagree with the truth state). We present a mathematical model for simulating binary MRMC data with a desired correlation structure across readers, cases, and two modalities, assuming the expected probability of agreement is equal for the two modalities ([Formula: see text]). This model can be used to validate the coverage probabilities of 95% confidence intervals (of [Formula: see text], [Formula: see text], or [Formula: see text] when [Formula: see text]), validate the type I error of a superiority hypothesis test, and size a noninferiority hypothesis test (which assumes [Formula: see text]). To illustrate the utility of our simulation model, we adapt the Obuchowski-Rockette-Hillis (ORH) method for the analysis of MRMC binary agreement data. Moreover, we use our simulation model to validate the ORH method for binary data and to illustrate sizing in a noninferiority setting. Our software package is publicly available on the Google code project hosting site for use in simulation, analysis, validation, and sizing of MRMC reader studies with binary agreement data.

eeDAP: An Evaluation Environment for Digital and Analog Pathology

PURPOSE

The purpose of this work is to present a platform for designing and executing studies that compare pathologists interpreting histopathology of whole slide images (WSI) on a computer display to pathologists interpreting glass slides on an optical microscope.

METHODS

Here we present eeDAP, an evaluation environment for digital and analog pathology. The key element in eeDAP is the registration of the WSI to the glass slide. Registration is accomplished through computer control of the microscope stage and a camera mounted on the microscope that acquires images of the real time microscope view. Registration allows for the evaluation of the same regions of interest (ROIs) in both domains. This can reduce or eliminate disagreements that arise from pathologists interpreting different areas and focuses the comparison on image quality.

RESULTS

We reduced the pathologist interpretation area from an entire glass slide (≈10-30 mm)² to small ROIs <(50 um)². We also made possible the evaluation of individual cells.

CONCLUSIONS

We summarize eeDAP's software and hardware and provide calculations and corresponding images of the microscope field of view and the ROIs extracted from the WSIs. These calculations help provide a sense of eeDAP's functionality and operating principles, while the images provide a sense of the look and feel of studies that can be conducted in the digital and analog domains. The eeDAP software can be downloaded from code.google.com (project: eeDAP) as Matlab source or as a precompiled stand-alone license-free application.

Confidence intervals for performance assessment of linear observers

PURPOSE

This work seeks to develop exact confidence interval estimators for figures of merit that describe the performance of linear observers, and to demonstrate how these estimators can be used in the context of x-ray computed tomography (CT). The figures of merit are the receiver operating characteristic (ROC) curve and associated summary measures, such as the area under the ROC curve. Linear computerized observers are valuable for optimization of parameters associated with image reconstruction algorithms and data acquisition geometries. They provide a means to perform assessment of image quality with metrics that account not only for shift-variant resolution and nonstationary noise but that are also task-based.

METHODS

We suppose that a linear observer with fixed template has been defined and focus on the problem of assessing the performance of this observer for the task of deciding if an unknown lesion is present at a specific location. We introduce a point estimator for the observer signal-to-noise ratio (SNR) and identify its sampling distribution. Then, we show that exact confidence intervals can be constructed from this distribution. The sampling distribution of our SNR estimator is identified under the following hypotheses: (i) the observer ratings are normally distributed for each class of images and (ii) the variance of the observer ratings is the same for each class of images. These assumptions are, for example, appropriate in CT for ratings produced by linear observers applied to low-contrast lesion detection tasks.

RESULTS

Unlike existing approaches to the estimation of ROC confidence intervals, the new confidence intervals presented here have exactly known coverage probabilities when our data assumptions are satisfied. Furthermore, they are applicable to the most commonly used ROC summary measures, and they may be easily computed (a computer routine is supplied along with this article on the Medical Physics Website). The utility of our exact interval estimators is demonstrated through an image quality evaluation example using real x-ray CT images. Also, strong robustness is shown to potential deviations from the assumption that the ratings for the two classes of images have equal variance. Another aspect of our interval estimators is the fact that we can calculate their mean length exactly for fixed parameter values, which enables precise investigations of sampling effects. We demonstrate this aspect by exploring the potential reduction in statistical variability that can be gained by using additional images from one class, if such images are readily available. We find that when additional images from one class are used for an ROC study, the mean AUC confidence interval length for our estimator can decrease by as much as 35%.

CONCLUSIONS

We have shown that exact confidence intervals can be constructed for ROC curves and for ROC summary measures associated with fixed linear computerized observers applied to binary discrimination tasks at a known location. Although our intervals only apply under specific conditions, we believe that they form a valuable tool for the important problem of optimizing parameters associated with image reconstruction algorithms and data acquisition geometries, particularly in x-ray CT.

New Theoretical Results on Channelized Hotelling Observer Performance Estimation with Known Difference of Class Means

Task-based assessments of image quality constitute a rigorous, principled approach to the evaluation of imaging system performance. To conduct such assessments, it has been recognized that mathematical model observers are very useful, particularly for purposes of imaging system development and optimization. One type of model observer that has been widely applied in the medical imaging community is the channelized Hotelling observer (CHO). Since estimates of CHO performance typically include statistical variability, it is important to control and limit this variability to maximize the statistical power of image-quality studies. In a previous paper, we demonstrated that by including prior knowledge of the image class means, a large decrease in the bias and variance of CHO performance estimates can be realized. The purpose of the present work is to present refinements and extensions of the estimation theory given in our previous paper, which was limited to point estimation with equal numbers of images from each class. Specifically, we present and characterize minimum-variance unbiased point estimators for observer signal-to-noise ratio (SNR) that allow for unequal numbers of lesion-absent and lesion-present images. Building on this SNR point estimation theory, we then show that confidence intervals with exactly-known coverage probabilities can be constructed for commonly-used CHO performance measures. Moreover, we propose simple, approximate confidence intervals for CHO performance, and we show that they are well-behaved in most scenarios of interest.

A nonparametric procedure for comparing the areas under correlated LROC curves

In contrast to the receiver operating characteristic (ROC) assessment paradigm, localization ROC (LROC) analysis provides a means to jointly assess the accuracy of localization and detection in an observer study. In a typical multireader, multicase (MRMC) evaluation, the data sets are paired so that correlations arise in observer performance both between readers and across the imaging conditions (e.g., reconstruction methods or scanning parameters) being compared. Therefore, MRMC evaluations motivate the need for a statistical methodology to compare correlated LROC curves. In this work, we suggest a nonparametric strategy for this purpose. Specifically, we find that seminal work of Sen on U-statistics can be applied to estimate the covariance matrix for a vector of LROC area estimates. The resulting covariance estimator is the LROC analog of the covariance estimator given by DeLong et al. for ROC analysis. Once the covariance matrix is estimated, it can be used to construct confidence intervals and/or confidence regions for purposes of comparing observer performance across imaging conditions. In addition, given the results of a small-scale pilot study, the covariance estimator may be used to estimate the number of images and observers needed to achieve a desired confidence interval size in a full-scale observer study. The utility of our methodology is illustrated with a human-observer LROC evaluation of three image reconstruction strategies for fan-beam x-ray computed tomography (CT).

Simulation tools for two-dimensional experiments in x-ray computed tomography using the FORBILD head phantom

Mathematical phantoms are essential for the development and early stage evaluation of image reconstruction algorithms in x-ray computed tomography (CT). This note offers tools for computer simulations using a two-dimensional (2D) phantom that models the central axial slice through the FORBILD head phantom. Introduced in 1999, in response to a need for a more robust test, the FORBILD head phantom is now seen by many as the gold standard. However, the simple Shepp-Logan phantom is still heavily used by researchers working on 2D image reconstruction. Universal acceptance of the FORBILD head phantom may have been prevented by its significantly higher complexity: software that allows computer simulations with the Shepp-Logan phantom is not readily applicable to the FORBILD head phantom. The tools offered here address this problem. They are designed for use with Matlab®, as well as open-source variants, such as FreeMat and Octave, which are all widely used in both academia and industry. To get started, the interested user can simply copy and paste the codes from this PDF document into Matlab® M-files.

On Efficient Assessment of Image-Quality Metrics Based on Linear Model Observers

This paper is motivated by the problem of image-quality assessment using model observers for the purpose of development and optimization of medical imaging systems. Specifically, we present a study regarding the estimation of the receiver operating characteristic (ROC) curve for the observer and associated summary measures. This study evaluates the statistical advantage that may be gained in ROC estimates of observer performance by assuming that the difference of the class means for the observer ratings is known. Such knowledge is frequently available in image-quality studies employing known-location lesion detection tasks together with linear model observers. The study is carried out by introducing parametric point and confidence interval estimators that incorporate a known difference of class means. An evaluation of the new estimators for the area under the ROC curve establishes that a large reduction in statistical variability can be achieved through incorporation of knowledge of the difference of class means. Namely, the mean 95% AUC confidence interval length can be as much as seven times smaller in some cases. We also examine how knowledge of the difference of class means can be advantageously used to compare the areas under two correlated ROC curves, and observe similar gains.

Line plus arc source trajectories and their R-line coverage for long-object cone-beam imaging with a C-arm system

Cone-beam imaging with C-arm systems has become a valuable tool in interventional radiology. Currently, a simple circular trajectory is used, but future applications should use more sophisticated source trajectories, not only to avoid cone-beam artifacts but also to allow extended volume imaging. One attractive strategy to achieve these two goals is to use a source trajectory that consists of two parallel circular arcs connected by a line segment, possibly with repetition. In this work, we address the question of R-line coverage for such a trajectory. More specifically, we examine to what extent R-lines for such a trajectory cover a central cylindrical region of interest (ROI). An R-line is a line segment connecting any two points on the source trajectory. Knowledge of R-line coverage is crucial because a general theory for theoretically exact and stable image reconstruction from axially truncated data is only known for the points in the scanned object that lie on R-lines. Our analysis starts by examining the R-line coverage for the elemental trajectories consisting of (i) two parallel circular arcs and (ii) a circular arc connected orthogonally to a line segment. Next, we utilize our understanding of the R-lines for the aforementioned elemental trajectories to determine the R-line coverage for the trajectory consisting of two parallel circular arcs connected by a tightly fit line segment. For this trajectory, we find that the R-line coverage is insufficient to completely cover any central ROI. Because extension of the line segment beyond the circular arcs helps to increase the R-line coverage, we subsequently propose a trajectory composed of two parallel circular arcs connected by an extended line. We show that the R-lines for this trajectory can fully cover a central ROI if the line extension is long enough. Our presentation includes a formula for the minimum line extension needed to achieve full R-line coverage of an ROI with a specified size, and also includes a preliminary study on the required detector size, showing that the R-lines added by the line extension are not constraining.

Band-restricted estimation of noise variance in filtered backprojection reconstructions using repeated scans

We introduce a new estimator for noise variance in tomographic images reconstructed using algorithms of the filtered backprojection type. The new estimator operates on data acquired from repeated scans of the object under examination, is unbiased, and is shown to have significantly lower variance than the conventional unbiased estimator for many scenarios of practical interest. We provide an extensive theoretical analysis of this estimator, highlighting the circumstances under which it is most effective. This analysis includes both general and specific data-correlation patterns. Moreover, we have applied our estimator to real X-ray computed tomography data and present preliminary results that support the theory and provide experimental evidence of the new estimator's efficacy.

Estimation of channelized hotelling observer performance with known class means or known difference of class means

This paper concerns task-based image quality assessment for the task of discriminating between two classes of images. We address the problem of estimating two widely-used detection performance measures, SNR and AUC, from a finite number of images, assuming that the class discrimination is performed with a channelized Hotelling observer. In particular, we investigate the advantage that can be gained when either 1) the means of the signal-absent and signal-present classes are both known, or 2) when the difference of class means is known. For these two scenarios, we propose uniformly minimum variance unbiased estimators of SNR(2), derive the corresponding sampling distributions and provide variance expressions. In addition, we demonstrate how the bias and variance for the related AUC estimators may be calculated numerically by using the sampling distributions for the SNR(2) estimators. We find that for both SNR(2) and AUC, the new estimators have significantly lower bias and mean-square error than the traditional estimator, which assumes that the class means, and their difference, are unknown.

Evaluation of the impact of tube current modulation on lesion detectability using model observers

We use several model observers to evaluate the influence of tube current modulation on lesion detectability in x-ray computed tomography images reconstructed using the classical direct fan-beam filtered backprojection algorithm. Specifically, we compute observer performance for a lesion detection task at various locations in an elliptic water cylinder using simulated x-ray data. The observer performance is computed both with and without x-ray tube current modulation. For different tube current modulation schemes, we find that the Hotelling observer and the channelized Hotelling observer with Gabor channels yield distinctly different qualitative behavior compared to channelized Hotelling observers with circularly symmetric channels and to non-prewhitening observers.

Image covariance and lesion detectability in direct fan-beam x-ray computed tomography

We consider noise in computed tomography images that are reconstructed using the classical direct fan-beam filtered backprojection algorithm, from both full- and short-scan data. A new, accurate method for computing image covariance is presented. The utility of the new covariance method is demonstrated by its application to the implementation of a channelized Hotelling observer for a lesion detection task. Results from the new covariance method and its application to the channelized Hotelling observer are compared with results from Monte Carlo simulations. In addition, the impact of a bowtie filter and x-ray tube current modulation on reconstruction noise and lesion detectability are explored for full-scan reconstruction.

Evaluation of image noise in fan-beam x-ray computed tomography

We investigate the problem of evaluating noise in computed tomography images that are reconstructed from x-ray measurements collected in fan-beam geometry. For reconstruction, we use the classical direct filtered backprojection method, which acts directly on the fan-beam projections without rebinning to the parallel-beam geometry. In this context, we present an accurate and efficient method to evaluate the image noise. This method is first validated using results from Monte-Carlo simulations. Then, we illustrate its advantages through the implementation of a channelized Hotelling observer for a lesion detection task.

Differential effects of cetuximab and AEE 788 on epidermal growth factor receptor (EGF-R) and vascular endothelial growth factor receptor (VEGF-R) in thyroid cancer cell lines

This study evaluated the role of EGF and the effects of EGF-targeting drugs (Cetuximab, AEE 788) on growth, apoptosis, and autocrine VEGF-secretion of thyroid cancer (TC) cells. Autocrine activation of the epidermal growth factor receptor (EGF-R) is commonly regarded to contribute to the malignant phenotype of TC cells and may therefore represent a rational therapeutic target. Out of a number of TC cell lines two anaplastic (Hth74, C643), one follicular (FTC133), and one papillary thyroid cancer cell line (TPC1) were analyzed in depth for VEGF-R-and EGF-R-expression, basal and EGF-stimulated (1-100 ng/ml) VEGF protein secretion and proliferation. Subsequently the antiprolifereative and antiangiogenic effect of cetuximab (Erbitux), a monoclonal antibody that blocks the EGF-R and AEE 788, a novel dual-kinase inhibitor of EGF-R and VEGF-R were assessed, and the downstream EGF-R signal transduction was analyzed by means of detecting phosphorylated pEGF-R, pVEGF-R, pAkt, and p-MAPK. EGF stimulated VEGF-mRNA expression and protein secretion in all TC cell lines. The EGF-R antagonist Cetuximab consistently decreased VEGF secretion in all TC cell lines (min. 15%, n.s. in C643 cells and max. 90% in Hth74 cells, P < 0.05), but did not affect tumor cell proliferation in vitro. In contrast, the EGF-R- and VEGF-R-kinase inhibitor AEE 788 not only reduced VEGF secretion (min. 55%, P < 0.05 in C643 and max. 75%, P < 0.05, in FTC133), but also exhibited a dose-dependent inhibition of tumor cell proliferation (min. 75%, P < 0.05 in C643 and max. 95%, P < 0.05 in Hth74) and was a potent inductor of apoptosis in two of four TC cell lines. These effects were always accompanied by reduced levels of pEGF-R, pVEGF-R, pAkt, and pMAPK. Although inhibition of the EGF-receptor by Cetuximab potently disrupts autocrine secretion of VEGF, only the concurrent inhibition of the VEGF- and EGF receptor, e.g., by AEE 788 induces reduced proliferation and apoptosis in vitro. This suggests a particular rationale for the use of tyrosine kinase inhibitors with dual modes of action such as AEE 788 in thyroid cancer.

Targeting the EGF/VEGF-R system by tyrosine-kinase inhibitors--a novel antiproliferative/antiangiogenic strategy in thyroid cancer

AIM

In thyroid cancer (TC), endothelial growth factor (EGF) has been associated with dedifferentiation, tumor cell proliferation, and angiogenesis. Vascular endothelial growth factor (VEGF) has been documented to be the main stimulator of angiogenesis in the thyroid gland. Patients with undifferentiated thyroid cancer are in desperate need of new therapeutic strategies because common protocols of therapy usually fail. The aim of this study, therefore, was to evaluate two tyrosine-kinase inhibitors (TKI, ZD 1839 gefitinib and ZD 6474 vandetanib), directed against the EGF/VEGF receptor for possible antitumor therapy in thyroid cancer.

METHODS

EGF/VEGF-R was documented in anaplastic (Hth74, C643), follicular (FTC133), and papillary (TPC1) thyroid cancer cell lines by Western blot analysis. The antiproliferative effect of two TKI (0.1-10 microM) on thyroid cancer cell lines in vitro was quantified by MTT assay, the antiangiogenic effect by assessing secretion of VEGF by enzyme-linked immunosorbent assay (R&D Systems). ZD 1839 is mainly directed against EGF-R and ZD 6474 against VEGF-R (AstraZeneca, UK), single applications and combinations of compounds were evaluated.

RESULTS

EGF-R and VEGF-R as well as the phosphorylated receptor were documented in all of the cell lines. Administration of ZD1839 led to an up to 90% reduction of cell number in Hth74, 80% in C643, 50% in FTC133, and 90% in TPC1 (p < 0.05). ZD1839 induced a decrease of VEGF secretion between 30% in C643 and 90% in Hth74. Administration of ZD6474 led to an up to 95% reduction of cell number in Hth74, 85% in C643, 90% in FTC133, and 90% in TPC1 (p < 0.05). The ZD6474 induced decrease of VEGF secretion ranged between 20% (FTC133) and 60% (TPC1). Combinations of IC50 concentrations of TKI showed synergistic effects, resulting in additional inhibition of proliferation between 50 and 90% compared to single drug administration.

CONCLUSION

The EGF/EGF-R system resembles a powerful VEGF-stimulating pathway in all histiotypes of TC and can be inhibited by TKI. TKI directed against EGF-R as well as VEGF-R inhibit tumor cell proliferation and VEGF secretion in vitro. Combinations of TKI are more effective than strategies using single agents. It is suggested that targeting EGF-R/VEGF-R-mediated pathways may have therapeutic potential in some undifferentiated thyroid cancers.

Functional thyrotropin receptor attenuates malignant phenotype of follicular thyroid cancer cells

Thyrotropin (TSH) is a thyroid-specific growth factor inducing differentiated function and growth of thyrocytes in vitro. In thyroid cancer, loss of TSH-receptor (TSHR) expression is a sign of de-differentiation and is believed to contribute to the malignant phenotype. The present studies aimed to determine the in vitro and in vivo effects of functioning TSHR in the follicular thyroid cancer cell line HTC, a subclone of FTC133 cells, lacking endogenous expression of TSHR, and HTCtshr+ cells transfected with human TSHR-cDNA. HTCtshr+ cells grew faster in vitro (doubling time 1.15 vs 1.56 d, p < 0.05) and TSH caused a dose-dependent growth response. Adhesion to and invasion through reconstituted basement membrane were reduced in HTCtshr+ cells, but when stimulated with TSH increased to levels comparable to naïve HTC cells. In vivo, tumor latency was 11 d for naïve HTC as compared to 21 d for HTCtshr+ xenografts. Smaller tumor volumes were registered for HTCtshr+ cells (250 +/- 217 vs 869 +/- 427 mm3, p < 0.05). Angiogenesis, as determined by vascular surface density (VSD) of experimental tumors, was enhanced in naïve HTC tumors (VSD 0.87 +/- 0.1 microm-1 vs 0.55 +/- 0.2 microm-1 in HTCtshr+, p < 0.05). VEGF secretion was more pronounced in naïve HTC cells stimulated with EGF, than in HTCtshr+ cells stimulated with either TSH or EGF. In conclusion, regained expression of functional TSHR in the follicular thyroid cancer cell line HTC alters in vitro features commonly associated with the malignant phenotype. Smaller tumors and reduced angiogenesis of xenotransplanted HTC cells with functioning TSHR suggest a less aggressive in vivo phenotype. The present data highlight the pivotal role of TSHR to affect transformed thyrocytes in vitro and in vivo. They also suggest a role for EGF as a modulator of angiogenesis in thyrocytes devoid of TSHR.

Brain tumour surgery in the vicinity of short-term memory representation--results of neuronavigation using fMRI images

OBJECTIVE

Functional information concerning the surrounding brain is mandatory for a good clinical outcome in brain tumour surgery. The value of fMRI to detect the motorcortex and Broca's area is widely accepted today. If an appropriate paradigm is used, short-term memory areas can be visualized as well. Obviously this information must be integrated into cranial neuronavigation for an appropriate intra-operative use. We report our first experiences with the direct integration of short-term memory fMRI into cranial neuronavigation.

METHOD

From January 2001 to March 2002 14 patients were operated on for intracranial tumours with short-term memory fMRI imaging, using the "two-back-paradigm". Both pre- and postoperatively, the short-term memory of all patients was tested additionally by a standardized test battery including 16 different verbal and visuo-spatial items.

RESULTS

In all 14 patients the general level of working memory capacity was preserved after surgery. The visuo-spatial performance was kept unchanged or deteriorated slightly, the alertness slightly worsened as well, but we found an improvement in verbal test items.

CONCLUSION

The two-back paradigm is able to visualize verbal memory tasks in fMRI. For visuo-spatial items, a new paradigm has to be designed. In contrast to deep seated brain lesions, focal cortical impairments do not lead to obvious and serious memory deficits. Therefore, the aim of gross total tumour removal has to be balanced against the aim of preservation of short-term memory fields. Nevertheless, the knowledge of the localization of cortical short-term memory fields may combine both aims during navigated brain tumour surgery, thus preserving the patient with a better quality of life.

Antisense p53 decreases production of VEGF in follicular thyroid cancer cells

Inactivating mutations of wild-type p53 (WTp53) tumor suppressor gene are common in anaplastic thyroid cancer (ATC) and are associated with poor prognosis. Mutated p53 (MTp53) has been implicated with angiogenesis. Therefore, the potential of MTp53 knockout by oligodeoxyribonucleotide phosphorothioates (ODNs) to affect VEGF production of undifferentiated thyroid cancer cells with a recessive MTp53 mutation was evaluated. Transient transfection with 20 bp ODNs complementary to portions of exon 10 of p53 and a negative control ODN (HIV-RT) were carried out in FTC-133 cells. In vitro secretion of VEGF protein was quantified by EIA and correlated to cell numbers, which was evaluated by in vitro MTT assay. Transfection of undifferentiated thyroid cancer cells with ODN reduced VEGF secretion of FTC-133 cells following transfection by 34% as compared to the negative control (cells transfected with ODN-HIV; p = 0.03). These results suggest that transient MTp53 knockout with ODNs complementary to p53 nucleotide sequences impair secretion of VEGF in the undifferentiated thyroid cancer cell line FTC-133.

Antisense p53 oligonucleotides inhibit proliferation and induce chemosensitivity in follicular thyroid cancer cells

BACKGROUND

The potential of MTp53 knockout by oligodesoxyribonucleotide phosphothioates (ODNs) to affect proliferation, apoptosis and chemosensitivity in undifferentiated thyroid cancer (UTC) cells with a recessive MTp53 mutation was evaluated.

MATERIALS AND METHODS

Transient transfections with ODNs complementary to p53 and control ODN (HIV-RT) were carried out in FTC 133 cells. In vitro proliferation was evaluated by cell counting of 10 random fields and by the MTT assay. A single pulse of 100 microg/ml Cytarabine was added to each well and the cells were incubated for an additional day. Chemosensitivity was calculated as the ratio of apoptotic and necrotic cells versus viable cells by flow cytometry (FACS).

RESULTS

Transfection of UTC cells with ODN decreased the cell number by up to 70% (p < 0.002). The proliferation rate also decreased up to 35% (p < 0.03), without inducing apoptosis. ODNs rendered FTC cells sensitive to treatment with Cytarabine, inducing apoptosis in 35% of cells, as compared to 17% of cells transfected with the reverse transcriptase gene of HIV (ODN-HIV) and less than 10% of non-transfected cells (p < 0.05).

CONCLUSION

Transient MTp53 knockout with ODNs complementary to p53 nucleotide sequences inhibited proliferation and increased chemosensitivity in the UTC cell line FTC133.

Internal radionuclide therapy: software for treatment planning using tomographic data

AIM

Accurate dosimetry must be performed for each patient before therapy with unsealed radionuclides. Recently, the software tool ULMDOS was developed to facilitate planar dosimetric calculations and to support traceability and documentation as a prerequisite for good clinical practice. Here, the extended version of ULMDOS for processing of tomographic data is presented.

METHODS

ULMDOS is developed in IDL 6.1 (Interactive Data Language) under Windows XP/2000. Serial tomographic data can be loaded in an ECAT7 or DICOM format, and presented as maximum intensity projection. The definition of volumes of interest is supported by various tools (e.g., freehand, isocontour, polygon), region growing, and cluster analysis. Residence times are calculated from fits of the time activity data to exponential functions.

RESULTS, DISCUSSION

Quantitative 3-dimensional data allow performing a more individualized dosimetry, as problems due to organ overlay, insufficient attenuation and scatter correction in the planar approach can be avoided. For traceability, documentation, retrospective examination and later processing all data can be saved in binary or ASCII format. Dosimetric calculations can be conducted within a single environment, thus it spares the time-consuming transfer of data between different software tools.

Differential pattern of integrin receptor expression in differentiated and anaplastic thyroid cancer cell lines

Adhesion of tumor cells to the extracellular matrix (ECM) is a crucial step for the development of metastatic disease and is mediated by specific integrin receptor molecules (IRM). The pattern of metastatic spread differs substantially among the various histotypes of thyroid cancer (TC). However, IRM have only occasionally been characterized in TC until now. IRM expression was investigated in 10 differentiated (FTC133, 236, 238, HTC, HTC TSHr, XTC, PTC4.0/4.2, TPC1, Kat5) and two anaplastic TC cell lines (ATC, C643, Hth74), primary cultures of normal thyroid tissue (Thy1,3), and thyroid cancer specimens (TCS). Expression of 16 IRM (beta1-4, beta7, alpha1-6, alphaV, alphaIIb, alphaL, alphaM, alphaX) and of four IRM heterodimers (alpha2beta1, alpha5beta1, alphaVbeta3, alphaVbeta5), was analyzed by fluorescent-activated cell sorter (FACS) and immunohistochemical staining. Thyroid tumor cell adhesion to ECM proteins and their IRM expression in response to thyrotropin (TSH) was assessed. Follicular TC cell lines presented high levels of integrins alpha2, alpha3, alpha5, beta1, beta3 and low levels of alpha1, whereas papillary lines expressed a heterogenous pattern of IRM, dominated by alpha5 and beta1. ATC mainly displayed integrins alpha2, alpha3, alpha5, alpha6, beta1 and low levels of alpha1, alpha4 and alphaV. Integrin heterodimers correlated with monomer expression. Evaluation of TCS largely confirmed these results with few exceptions, namely alpha4, alpha6, and beta3. The ability of TC cell lines to adhere to purified ECM proteins correlated with IRM expression. TSH induced TC cell adhesion in a dose-dependent fashion, despite an unchanged array of IRM expression or level of a particular IRM. Thyroid carcinoma cell lines of different histogenetic background display profoundly different patterns of IRM expression that appear to correlate with tumor aggressiveness. In vitro adhesion to ECM proteins and IRM expression concur. Finally, TSH-stimulated adhesion of thyroid tumor cell lines to ECM may not be associated with altered IRM expression.

Messungen zum Lärmschutz bei der funktionellen MR-Tomographie

BACKGROUND

Functional magnetic resonance imaging (fMRI) can detect changes in oxygen saturation of the brain. Fast changing high gradient fields are necessary which produce high levels of noise. In studies of the auditory cortex, auditory stimuli have to be perceived and discriminated against the noise level of the activated tomograph.

MATERIAL AND METHODS

The generated frequency bands and their intensities during fMRI with a Siemens Magnetom Vision, 1.5 T, EPI sequence were measured in the outer ear canal of a dummy head. Noise attenuation was evaluated with four different noise muffs (simple/inexpensive products, quality product, specialized fMRI muffs).

RESULTS

Without protection, peak noise levels reached up to 111 dB(A) near 1000 Hz in the dummy ear canal. Major noise attenuation was only found at higher frequencies (4000 Hz by about 25 dB; 8000 Hz by about 35 dB) with the quality product and the specialized fMRI muffs.

CONCLUSION

Only quality noise products can sufficiently protect patients from high sound pressure levels of tomograph noise. If in the future higher gradient fields are applied at faster slew rates, acoustic stimuli can safely be applied only in combination with increased hearing protection systems in order to minimize the risk of noise trauma.

Quasi-automatic 3D finite element model generation for individual single-rooted teeth and periodontal ligament

The paper demonstrates how to generate an individual 3D volume model of a human single-rooted tooth using an automatic workflow. It can be implemented into finite element simulation. In several computational steps, computed tomography data of patients are used to obtain the global coordinates of the tooth's surface. First, the large number of geometric data is processed with several self-developed algorithms for a significant reduction. The most important task is to keep geometrical information of the real tooth. The second main part includes the creation of the volume model for tooth and periodontal ligament (PDL). This is realized with a continuous free form surface of the tooth based on the remaining points. Generating such irregular objects for numerical use in biomechanical research normally requires enormous manual effort and time. The finite element mesh of the tooth, consisting of hexahedral elements, is composed of different materials: dentin, PDL and surrounding alveolar bone. It is capable of simulating tooth movement in a finite element analysis and may give valuable information for a clinical approach without the restrictions of tetrahedral elements. The mesh generator of FE software ANSYS executed the mesh process for hexahedral elements successfully.

Verbal storage in a premotor-parietal network: evidence from fMRI-guided magnetic stimulation

Phonological rehearsal helps to keep selected information consciously in mind for further processing. This part of short-term storage takes place during the delay period of verbal working memory tasks and involves a frontoparietal network as functional magnetic resonance imaging (fMRI) studies have shown. The involved cortical areas can be further investigated by interfering with the local information processing using transcranial magnetic stimulation (TMS). On a single subject level, we identified predominantly left-sided premotor, prefrontal, and parietal areas active during the delay period of a verbal working memory task using event-related fMRI. In a pilot approach, TMS was neuronavigated to the individually active areas by using a stereotaxic device. Then, TMS was applied during the delay period of similar tasks as in fMRI. Error rates increased significantly upon stimulating left premotor cortex, but not upon parietal or prefrontal stimulation. The contribution of the premotor cortex to storage and rehearsal is discussed as an active top-down storage process within the frontoparietal network.

Der Hirnabszess in der Diffusionswichtung und Unterschiede zum zystischen Hirntumor

PURPOSE

The clinical usefulness of diffusion-weighted imaging (DWI) was evaluated in patients with brain abscesses in comparison to patients with cystic brain tumors.

MATERIAL AND METHODS

Five patients with surgically confirmed brain abscesses underwent beside a brain MRI examination with contrast media application diffusion weighted imaging. Apparent diffusion coefficients (rADC) in three orthogonal diffusion gradient were calculated. The same protocol was used to examine 5 patients with cystic brain tumors.

RESULTS

Showing an rADC of 0.33 x 10(-3)/mm(2)/s abscesses have a highly restricted diffusion in comparison to cystic brain tumors with an rADC of 1,67 x 10(-3)/mm(2)/s.

CONCLUSION

Diffusion weighted imaging is a usefull diagnostic tool in the work up of brain abscesses.

Practicability and acute haematological toxicity of 2- and 3-weekly CHOP and CHOEP chemotherapy for aggressive non-Hodgkin's lymphoma: results from the NHL-B trial of the German High-Grade Non-Hodgkin's Lymphoma Study Group (DSHNHL)

BACKGROUND

There is evidence that intensified variants of the classical 3-weekly CHOP-21 chemotherapy [cyclophosphamide (C), doxorubicin (H), vincristine (O), prednisone (P)] may improve treatment outcome in aggressive lymphoma. Three variants using either an addition of etoposide (CHOEP-21: 100 mg/m(2) on days 1-3), the shortening to 2-week intervals using recombinant human granulocyte colony-stimulating factor (rhG-CSF; CHOP-14) or both (CHOEP-14) are currently compared with CHOP-21 in the NHL-B trial of the German High-Grade Non-Hodgkin's Lymphoma Study Group (DSHNHL). To enable more extensive testing of these schemes we here characterise their practicability regarding schedule adherence, acute haematotoxicity and need for supportive treatment.

PATIENTS AND METHODS

The trial included patients with normal lactate dehydrogenase (LDH) aged </=60 years (NHL-B1) and patients aged 61-75 years (NHL-B2). The data are taken from an interim analysis. Data from 959 patients (CHOP-21: 232; CHOP-14: 238; CHOEP-21: 244; CHOEP-14: 245) from 162 institutions with a total of 5331 therapy cycles were evaluated.

RESULTS

The dose adherence in the NHL-B1 trial was excellent. The median relative dose (RD; i.e. actually given compared to planned dose) exceeds 98% for the myelosuppressive drugs in all four regimens. Only </=5% of patients received a relative dose <80% (RD <80). The median treatment duration could be shortened as scheduled for both CHOP-14 by 36 days and CHOEP-14 by 35 days. The dose adherence in the NHL-B2 trial was excellent for CHOP-21 and CHOP-14 for the myelosuppressive drugs (median RD >/=98%, RD <80 </=15%). Addition of etoposide, however, was accompanied by more dose erosion (median RD >/=97%, RD <80 </=17% for CHOEP-21 and </=27% for CHOEP-14). The median treatment duration could be shortened by 34 days with CHOP-14 compared with CHOP-21. Less treatment shortening was feasible for CHOEP-14 compared with CHOP-21 (median of 29 days). CHOP-14 and CHOP-21 were similar regarding toxicity profile, rate of infection, use of antibiotics, rate of transfusions and hospitalisation. CHOEP schemes were associated with a higher rate of infections, more transfusion requirements, more antibiotic use and longer hospitalisation than the CHOP schemes, particularly in patients aged >60 years. Haematopoietic recovery was age- and treatment-related.

CONCLUSIONS

CHOP-14 with the addition of rhG-CSF is safe and practicable in a large multicentre setting in patients aged 18-75 years. Despite shorter treatment intervals it can be delivered at the same dose as the classical 3-weekly CHOP with a comparable toxicity profile. The addition of etoposide is feasible and safe for patients </=60 years old in both the CHOEP-21 and CHOEP-14 schemes. For patients >60 years of age the addition of etoposide is associated with marked dose erosion due to increased toxicity. In this age group CHOEP should be used with caution.

In vitro and in vivo angiogenesis in PC12 pheochromocytoma cells is mediated by vascular endothelial growth factor

Angiogenesis, the formation of new blood vessels from existing vascular endothelium, is essential for tumor growth. Vascular endothelial growth factor (VEGF) is an endotheliumspecific mitogen and regulator of angiogenesis. Angiogenesis has been associated to the malignant phenotype of pheochromocytomas and is readily observed in experimental pheochromocytomas. Although VEGF gene expression has already been demonstrated in the rat PC12 cell line, the detailed mechanisms of action are not known. We have, therefore, studied angiogenesis in the rat PC12 pheochromocytoma cell line in vitro and in vivo. VEGF gene expression and accumulation of VEGF protein in cytoplasm and conditioned medium of PC12 cells was found. Conditioned medium from PC12 cells significantly increased proliferation of VEGF-dependent endothelial cells from human umbilical veins, and this effect reversed upon addition of a neutralizing anti-VEGF antibody. Dexamethasone and nerve growth factor (NGF) increased VEGF mRNA expression and accumulation of VEGF protein of PC12 subclones with established metastatic activity in vivo. PC12 cells xenotransplanted to nude mice had marked VEGF expression and induced host angiogenesis, confirmed by the presence of CD34-positive endothelial cells in the experimental PC12 tumors. When NGF-primed PC12 cells were immobilized in Matrigel supplemented with rising concentrations of the growth factor and xenotransplanted, increasing NGF resulted in tumors with smaller areas of necrosis and increased vital tumor volume. These results suggest that VEGF is a mediator of angiogenesis in the PC12 pheochromocytoma cell line, and that dexamethasone and NGF affect VEGF expression. Our data further suggest that NGF may contribute to angiogenesis in experimental pheochromocytoma.

Multimodal cranial neuronavigation: direct integration of functional magnetic resonance imaging and positron emission tomography data: technical note

OBJECTIVE

This is the first report of the direct integration of functional magnetic resonance imaging (fMRI) and positron emission tomography (PET) data into cranial neuronavigation.

METHODS

In a patient with a left precentral oligodendroglioma (World Health Organization Grade III), the Zeiss MKM system (Carl Zeiss Co., Oberkochen, Germany) was used for navigation based on thin-slice, T1-weighted, contrast-enhanced magnetic resonance imaging (MRI) scans. fMRI and methionine PET data were integrated by landmark matching, with reference to skin fiducials.

RESULTS

The inaccuracy of the image fusion between fMRI and T1-weighted MRI data was 1.7 mm, that between PET and T1-weighted MRI data was 4.3 mm, and that for the subsequent registration of the navigation was 1.2 mm. The correct fMRI localization of the precentral gyrus was intraoperatively verified by cortical somatosensory evoked potential (phase-reversal) monitoring. Although the tumor was not clearly defined in the MRI scans, [11C]methionine PET demonstrated a clear tumor border, enabling us to achieve gross total tumor removal without postoperative functional deficits.

CONCLUSION

Functional neuronavigation permits observation and preservation of relevant brain areas. Other functional areas (such as short-term memory areas) that can be detected only by fMRI might also warrant future monitoring. The simultaneous integration of fMRI and PET data adds a new dimension to cranial neuronavigation, enabling the observation of tumors in relation to functional cortical areas (in our case, the motor strip).

Diffusion-weighted MRI: detection of cerebral ischemia before and after carotid thromboendarterectomy

PURPOSE

Conventional postoperative evaluation of patients following carotid thromboendarterectomy (TEA) consists of a clinical neurologic examination to assess neurologic deficits, color duplex ultrasound to document the surgically reestablished patency of the carotid artery, and CT for exclusion of postoperative ischemic infarctions. Recent studies prove that diffusion-weighted MRI is more sensitive in the detection of fresh insults than conventional MRI and CT. The objective of the study was to ascertain the incidence of clinically asymptomatic peri-and postoperative ischemic infarctions visualized at MRI.

METHOD

We included 52 patients in the study. Fifty-one patients (31 men, 20 women; average age 68 years) underwent cranial MR examination including a diffusion-weighted sequence at 24 h prior to carotid TEA and again 24 h following the procedure. One patient did not agree to participate.

RESULTS

In 29 of 51 patients (56%), neither the pre-nor the postoperative MR scans showed any diffusion abnormalities. In 16 patients (31%), however, preoperative MRI detected fresh ischemic insults. In nine patients (17.6%), the size of the insult resulted in surgery being postponed for 4 weeks. In six patients (11.8%), postoperative MRI returned findings of fresh disturbances of diffusion suggestive of ischemia that were not visualized on preoperative scans. Discrete neurologic deficits were observed in only two (3.9%) of these patients. Deficits were transient and disappeared within 72 h.

CONCLUSION

Our findings underscore MRI's capacity for visualizing perioperative ischemic events. Moreover, MRI provides evidence of clinically asymptomatic embolisms that occur perioperatively.

Functional cranial neuronavigation. Direct integration of fMRI and PET data

OBJECTIVE

We report our first experiences with the direct integration of fMRI data into cranial neuronavigation.

METHOD

For navigation we used the MKM system and thin-sliced T1 contrast enhanced images. As a first step 21 patients had fMRI for localization of the precentral gyrus, 2 patients for Broca area detection. By anatomical correlation, these functional data were indirectly compared to the intraoperative findings using cortical SSEP (n=20) or cortical stimulation (n=3). Encouraged by these preliminary results, we started the direct integration of fMRI into neuronavigation in June 1999, followed by PET in January 2000, enabling us to compare functional images with intraoperative findings directly. fMRI and PET data were integrated by landmark matching referring on skin fiducials. Meanwhile, fMRI data of 8 patients (6 motorcortex, 2 Broca) and PET images of 1 patient were directly integrated into neuronavigation. Six out of 8 patients had additional cortical monitoring, 2/8 were exclusively operated on by functional neuronavigation.

RESULTS

Using indirect comparison between fMRI and intraoperative findings we observed a good correlation in every case for the motorcortex, but only in 1/2 for the speech area. In all 6 direct integrated fMRI cases, these findings corresponded well to the conventional ones. Both patients with sole functional navigation did not have any postoperative neurological deficit. The inaccuracy of the fMRI ifT1 matching was 2. 7 mm (sigma=0.9 mm) and 1.3 mm (sigma=0.4 mm) of the subsequent referenciation of the navigation. The tumor delinement shown by 11C-methionine PET could be proven by intraoperative biopsy outside its indicated tumor margin. The inaccuracy of the PET matching was 0. 8 mm.

CONCLUSION

Functional neuronavigation enables to visualize and preserve relevant brain areas. Other functional areas like short-term memory, which solely can be detected by fMRI might also be monitored in the future. The integration of PET data expect to gain a better differentiation of tumor and edema.