Statistical validation of image segmentation quality based on a spatial overlap index

RATIONALE AND OBJECTIVES: To examine a statistical validation method based on the spatial overlap between two sets of segmentations of the same anatomy. MATERIALS AND METHODS: The Dice similarity coefficient (DSC) was used as a statistical validation metric to evaluate the performance of both the reproducibility of manual segmentations and the spatial overlap accuracy of automated probabilistic fractional segmentation of MR images, illustrated on two clinical examples. Example 1: 10 consecutive cases of prostate brachytherapy patients underwent both preoperative 1.5T and intraoperative 0.5T MR imaging. For each case, 5 repeated manual segmentations of the prostate peripheral zone were performed separately on preoperative and on intraoperative images. Example 2: A semi-automated probabilistic fractional segmentation algorithm was applied to MR imaging of 9 cases with 3 types of brain tumors. DSC values were computed and logit-transformed values were compared in the mean with the analysis of variance (ANOVA). RESULTS: Example 1: The mean DSCs of 0.883 (range, 0.876-0.893) with 1.5T preoperative MRI and 0.838 (range, 0.819-0.852) with 0.5T intraoperative MRI (P < .001) were within and at the margin of the range of good reproducibility, respectively. Example 2: Wide ranges of DSC were observed in brain tumor segmentations: Meningiomas (0.519-0.893), astrocytomas (0.487-0.972), and other mixed gliomas (0.490-0.899). CONCLUSION: The DSC value is a simple and useful summary measure of spatial overlap, which can be applied to studies of reproducibility and accuracy in image segmentation. We observed generally satisfactory but variable validation results in two clinical applications. This metric may be adapted for similar validation tasks.     

Volumetric assessment of tumor infiltration of adjacent white matter based on anatomic MRI and diffusion tensor tractography

RATIONALE AND OBJECTIVES: To perform a retrospective, quantitative assessment of the anatomic relationship between intra-axial, supratentorial, primary brain tumors, and adjacent white matter fiber tracts based on anatomic and diffusion tensor magnetic resonance imaging (MRI). We hypothesized that white matter infiltration may be common among different types of tumor. MATERIAL AND METHODS: Preoperative, anatomic (T1- and T2-weighted), and LINESCAN diffusion tensor MRI were obtained in 12 patients harboring supratentorial gliomas (World Health Organization [WHO] Grades II and III). The two imaging modalities were rigidly registered. The tumors were manually segmented from the T1- and T2-weighted MRI, and their volume calculated. A three-dimensional tractography was performed in each case. A second segmentation and volume measurement was performed on the tumor regions intersecting adjacent white matter fiber tracts. Statistical methods included summary statistics to examine the fraction of tumor volume infiltrating adjacent white matter. RESULTS: There were five patients with low-grade oligodendroglioma (WHO Grade II), one with low-grade mixed oligoastrocytoma (WHO Grade II), one with ganglioglioma, two with low-grade astrocytoma (WHO Grade II), and three with anaplastic astrocytoma (WHO Grade III). We identified white matter tracts infiltrated by tumor in all 12 cases. The median tumor volume (+/- standard deviation) in our patient population was 42.5 +/- 28.9 mL. The median tumor volume (+/- standard deviation) infiltrating white matter fiber tracts was 5.2 +/- 9.9 mL. The median percentage of tumor volume infiltrating white matter fiber tracts was 21.4% +/- 9.7%. CONCLUSIONS: The information provided by diffusion tensor imaging combined with anatomic MRI might be useful for neurosurgical planning and intraoperative guidance. Our results confirm previous reports that extensive white matter infiltration by primary brain tumors is a common occurrence. However, prospective, large population studies are required to definitively clarify this issue, and how infiltration relates to histologic tumor type, tumor size, and location.     

Lung tumor growth: assessment with CT--comparison of diameter and cross-sectional area with volume measurements

PURPOSE: To compare diameter and cross-sectional area measurements with volume measurements in the assessment of lung tumor growth with serial computed tomography (CT). MATERIALS AND METHODS: Patients with lung cancer who underwent at least one pair of chest CT examinations 25 or more days apart before treatment and with a tumor size of T1 (< or =3-cm diameter) at the initial CT examination were identified. A total of 63 patients (62 men, one woman) who underwent 93 pairs of CT examinations were included. Images obtained at each examination were displayed, and the maximum diameter, cross-sectional area, and volume of the tumor were measured. For each measurement, the change between examinations was assessed to determine whether the change reached a detection threshold for growth, as determined in a prior study with simulated tumors. Results were then compared between measurement methods, with volume change serving as the reference standard, by calculating Spearman rank-order coefficients between examinations. Tumor size or section width were also evaluated with the two-tailed Fisher exact probability test to determine if they affected agreement about tumor growth between measurement methods. RESULTS: Thresholds were as follows: diameter, 2.1 mm with hand-held calipers and 0.68 mm with electronic calipers; area, 9.4%; volume, 16.5%. The median time between examinations was 92 days (range, 25-1,221 days). Median diameter increased from 19.3 mm to 23.0 mm (19.2%), median area from 207 mm(2) to 267 mm(2) (29.0%), and median volume from 1,652 mm(3) to 2,443 mm(3) (47.9%). Growth assessment with these diameter (as assessed with hand-held and electronic calipers) and area thresholds disagreed with those obtained with volume in 34 (37%), 26 (28%), and 25 (27%) of the 93 pairs of CT examinations, respectively. Of diameter assessments with the hand-held caliper threshold, 28 (30%) were false-negative; false-negative results occurred with this diameter threshold and area threshold with examination intervals as long as 1 year. CONCLUSION: Growth assessment of T1 lung tumors on serial CT scans with nonvolumetric measurements frequently disagrees with growth assessment with volumetric measurements.     

Semiautomatic nonrigid registration for the prostate and pelvic MR volumes

RATIONALE AND OBJECTIVES: Three-dimensional (3D) nonrigid image registration for potential applications in prostate cancer treatment and interventional magnetic resonance (iMRI) imaging-guided therapies were investigated. MATERIALS AND METHODS: An almost fully automated 3D nonrigid registration algorithm using mutual information and a thin plate spline (TPS) transformation for MR images of the prostate and pelvis were created and evaluated. In the first step, an automatic rigid body registration with special features was used to capture the global transformation. In the second step, local feature points (FPs) were registered using mutual information. An operator entered only five FPs located at the prostate center, left and right hip joints, and left and right distal femurs. The program automatically determined and optimized other FPs at the external pelvic skin surface and along the femurs. More than 600 control points were used to establish a TPS transformation for deformation of the pelvic region and prostate. Ten volume pairs were acquired from three volunteers in the diagnostic (supine) and treatment positions (supine with legs raised). RESULTS: Various visualization techniques showed that warping rectified the significant pelvic misalignment by the rigid-body method. Gray-value measures of registration quality, including mutual information, correlation coefficient, and intensity difference, all improved with warping. The distance between prostate 3D centroids was 0.7 +/- 0.2 mm after warping compared with 4.9 +/- 3.4 mm with rigid-body registration. CONCLUSION: Semiautomatic nonrigid registration works better than rigid-body registration when patient position is changed greatly between acquisitions. It could be a useful tool for many applications in the management of prostate.     

Robust dynamic susceptibility contrast MR perfusion using 4D nonlinear noise filters

PURPOSE: To investigate if 4D (simultaneous space and time) nonlinear filtering techniques can produce more robust cerebral blood flow (CBF) estimates by reducing noise in acquired dynamic susceptibility contrast (DSC) MR perfusion data. MATERIALS AND METHODS: A digital anthropomorphic brain perfusion phantom was constructed to analyze filter performance by: 1) deriving anthropomorphic tissue volume fractions from a human subject and 2) simulating DSC-MR perfusion signals for voxels with mixed tissue for various signal-to-noise ratios (SNRs). DSC-MR data for 11 acute ischemic stroke patients were also acquired at 3T. CBF maps cross-calibrated so that normal white matter CBF was 22 mL/minute/100 g were produced from DSC-MR data without filtering and from 4D-Gaussian and 4D-bilateral noise-filtered DSC-MR data. RESULTS: The nonlinear 4D-bilateral filter yielded the lowest CBF root-mean square error (RMSE) in the phantom experiments with noise (average RMSE across all tissues regions for no filtering, 4D-Gaussian, and 4D-bilateral was 5.3 mL/minute/100 g, 6.2 mL/minute/100 g, and 4.0 mL/minute/100 g, respectively) and had the best image quality in both the phantom and patient data. CONCLUSION: Nonlinear 4D noise filters are better suited to the 4D nature of DSC-MR data. Linear spatial filters are not appropriate and can produce larger CBF errors than without filtering.     

Visualization of MR angiographic data with segmentation and volume-rendering techniques.

Novel image processing and computer graphics techniques were developed to create three-dimensional (3D) models of vasculature from magnetic resonance (MR) angiographic images of the head or neck. Region growing was used to produce a mask that isolated the vascular signal in the MR angiographic data. The masked images were subjected to gradient-shaded volume rendering to create 3D views of the vasculature. The computer-derived model of intracranial vasculature was then merged with a 3D model of brain parenchyma derived from a set of MR images. The combined display of vascular and gyral anatomy may be useful for neurosurgical planning.     

Cerebral edema attenuated inversion recovery MR sequence in low magnetic field: a feasibility study

RATIONALE AND OBJECTIVES: Minimally invasive neurosurgery requires methods to specify surgical boundaries of target tissue, such as brain tumors. This study investigated technical possibilities and clinical usefulness of adapting edema attenuated inversion recovery (EDAIR) pulse sequences to suppress magnetic resonance signal from cerebral edema in brain tumor patients. MATERIALS AND METHODS: A resistive 0.23-T magnetic resonance scanner with magnitude-encoded inversion recovery sequences was used. Twenty-eight separate scanning tests in 25 neurosurgical brain tumor patients were performed on the day before surgery. An inversion recovery sequence with several inversion times between 150 and 2,200 ms was tested. The same sequences were also used intraoperatively and postoperatively. RESULTS: T(1) relaxation time of brain edema varied from case to case. An inversion recovery sequence with an inversion time of 400-800 milliseconds attenuated brain edema and seemed to help in demarcating gross brain tumor for surgical resection. These features were helpful for the evaluation of resectable tumor tissue particularly using neuronavigation techniques. CONCLUSIONS: According to these preliminary findings, inversion recovery sequences supplement other imaging modalities and assist neurosurgeons in evaluating different surgical trajectories and in estimating brain tumor volume before craniotomy.     

Proton Beam Therapy for Hepatocellular Carcinoma Patients with Severe Cirrhosis

BACKGROUND AND PURPOSE: Hepatocellular carcinoma (HCC) patients with severe cirrhosis are usually treated with supportive care because of their poor prognosis. However, the survival of severe cirrhotic patients has recently improved due to advanced treatments. The aim of this study was to define the role of proton beam therapy for HCC patients with severe cirrhosis. PATIENTS AND METHODS: 19 HCC patients with Child-Pugh class C cirrhosis received proton beam therapy. The hepatic tumors were solitary in 14 patients and multiple in five, and the tumor size was 25-80 mm (median 40 mm) in maximum diameter. No patient had regional lymph node or distant metastasis. Total doses of 50-84 Gy (median 72 Gy) in ten to 24 fractions (median 16) were delivered to the tumors. RESULTS: Of the 19 patients, six, eight and four died of cancer, liver failure and intercurrent diseases, respectively, during the follow-up period of 3-63 months (median 17 months) after treatment. A remaining patient was alive with no evidence of disease 33 months after treatment. All but one of irradiated tumors were controlled during the follow-up period. Ten patients had new intrahepatic tumors outside the irradiated volume. The overall and progression-free survival rates were 53% and 47% at 1 year, respectively, and 42% each at 2 years. Performance status and Child-Pugh score were significant prognostic factors for survival. Therapy-related toxicity of grade 3 or more was not observed. CONCLUSION: Proton beam therapy for HCC patients with severe cirrhosis was tolerable. It may improve survival for patients with relatively good general condition and liver function.     

Fully automatic segmentation of the brain from T1-weighted MRI using Bridge Burner algorithm

PURPOSE: To validate Bridge Burner, a new brain segmentation algorithm based on thresholding, connectivity, surface detection, and a new operator of constrained growing. MATERIALS AND METHODS: T1-weighted MR images were selected at random from three previous neuroimaging studies to represent a spectrum of system manufacturers, pulse sequences, subject ages, genders, and neurological conditions. The ground truth consisted of brain masks generated manually by a consensus of expert observers. All cases were segmented using a common set of parameters. RESULTS: Bridge Burner segmentation errors were 3.4% +/- 1.3% (volume mismatch) and 0.34 +/- 0.17 mm (surface mismatch). The disagreement among experts was 3.8% +/- 2.0% (volume mismatch) and 0.48 +/- 0.49 mm (surface mismatch). The error obtained using the brain extraction tool (BET), a widely used brain segmentation program, was 8.3% +/- 9.1%. Bridge Burner brain masks are visually similar to the masks generated by human experts. Areas affected by signal intensity nonuniformity artifacts were occasionally undersegmented, and meninges and large sinuses were often falsely classified as the brain tissue. Segmentation of one MRI dataset takes seven seconds. CONCLUSION: The new fully automatic algorithm appears to provide accurate brain segmentation from high-resolution T1-weighted MR images.     

The normal neonatal brain: MR imaging, diffusion tensor imaging, and 3D MR spectroscopy in healthy term neonates

BACKGROUND AND PURPOSE: There is a lack of normative diffusion tensor imaging (DTI) and 3D MR spectroscopy (MRS) data in the early neonatal period. We report quantitative values from a cohort of healthy term neonates to serve as baseline data for studies assessing brain development and injury. MATERIALS AND METHODS: Sixteen healthy term neonates (median age, 7 days) were studied with spin-echo T1- and T2-weighted MR imaging, DTI, and 3D point-resolved spectroscopy sequence (PRESS) MRS without sedation on a 1.5 T scanner. Average diffusivity (D(av)), fractional anisotropy (FA), eigenvalues (EV), and metabolite ratios (N-acetylaspartate [NAA]/choline, lactate/choline) were calculated by automated processing in 7 brain regions. Neurodevelopment was assessed by blinded and validated neuromotor examinations and the Bayley II test at 3 and 14 months. RESULTS: Two neonates were excluded from the cohort: one had brain injury on T2-weighted imaging, and the other, who had normal MR imaging, showed mildly delayed cognition at 14 months. The mean DTI values of the remaining 14 neonates were between these ranges: D(av)=0.98-1.48 10(-3) mm(2)/s, FA=0.14-0.30, EV1=1.21-1.88, EV2=0.95-1.46, and EV3=0.77-1.24 (all x 10(-3) mm(2)/s). The NAA/choline ratio ranged between 0.58 and 0.73, and minimal lactate/choline (<0.15) could be detected in each neonate. All neonates exhibited clinically normal neuromotor status. CONCLUSIONS: Our study demonstrates the feasibility of obtaining high-quality quantifiable MR data in nonsedated healthy term neonates that can be used to study normal early brain development and as control data in studies of perinatal brain injury.     

MRI evaluation and measurement of the normal odontoid peg position

AIM: To measure the normal distances (and range) from the tip of the odontoid peg to the different reference skull baselines (Chamberlain's, McGregor's, and McRae's line) using magnetic resonance imaging (MRI). MATERIALS AND METHODS: We retrospectively evaluated midline sagittal MRI brain images of 200 adults chosen randomly. Patients did not have symptoms or signs suggestive of basilar impression, spinal trauma, vertebral collapse or disease. Using SPSS data analysis program histograms, mean and standard deviation (SD), median and range values were calculated. These findings were then compared with previous plain radiograph measurements. RESULTS: The mean position of the odontoid peg was 1.2 mm (median 1.5 mm, SD 3 mm) below Chamberlain's line; 0.9 mm (median 1.1, SD 3 mm) below McGregor's line; and 4.6 mm (median 4.8, SD 2.6) below McRae's line. CONCLUSION: Based on the current population, these results provide the mean and range of normal distances from the odontoid peg to the most frequently used skull baselines using MRI.     

Does the Imaging Method Have an Influence on the Measured Tumor Height in Ruthenium Plaque Therapy of Uveal Melanoma?

BACKGROUND AND PURPOSE: For ruthenium plaque therapy of uveal melanoma the precise knowledge of the tumor height is necessary. The purpose of this analysis is to examine systematic differences between ultrasound (US), computed tomography (CT), and magnetic resonance imaging (MRI) measurements of tumor height in uveal melanoma. MATERIAL AND METHODS: From 1993 to 2000, 208 patients with uveal melanoma were treated with ruthenium therapy. US was done in 190/208 patients, CT in 147/208 patients, and MRI in 81/208 patients. The measurements from the imaging modalities were compared to each other. Separate comparisons were carried out for tumor size and localization and between two-dimensional (2-D) and three-dimensional (3-D) measurements in reconstructed planes. RESULTS: In direct comparison, CT measurements (median: 4.5 mm; range: 1.6-10.5 mm) were nearly identical to MRI measurements (median: 4.5 mm; range: 2.0-11.4 mm), while US yielded the highest values (median: 5.2 mm; range: 2.5-13.4 mm). Linear regression yielded the following values: CT versus US (R(2) = 0.88, correlation coefficient = 1.04), MRI versus US (R(2) = 0.79, correlation coefficient = 0.92), MRI versus CT (R(2) = 0.84, correlation coefficient = 0.90), and "2-D" versus reconstructed "3-D" (R2 = 0.93, correlation coefficient = 0.98). CONCLUSION: Differences between the measurements from all three modalities were generally acceptable, except for small tumors (< or =4 mm; limited spatial resolution of CT/MRI) and temporal or peripheral lesions (angular distortion in US). In special anatomic situations, CT/MRI measurements can also result in distortions. "3-D" measurements can sometimes help in these situations.     

Human brain hemorrhage: quantification of perihematoma edema by use of diffusion-weighted MR imaging

BACKGROUND AND PURPOSE: Animal models have clearly shown a critical role for extravascular blood in the initiation of the vasogenic edema associated with intracerebral hemorrhage (ICH). Nevertheless, the relevance of these observations to the human disease process has not been evaluated. With a prospectively collected cohort of nine patients, we report the relation between intraparenchymal blood clot volume and elevation of perihematoma brain tissue (and homologous contralateral brain tissue) apparent diffusion coefficient (ADC). METHODS: Patients with acute and subacute supratentorial ICH were prospectively evaluated by using diffusion-weighted imaging. ADC was measured in perihematoma tissue and in homologous contralateral regions. The relationship between ADC and volume of hematoma was determined by using linear regression analysis. RESULTS: Nine patients were enrolled in the study. The mean hematoma volume was 30.8 cc (range, 2.6-74 cc). The ADC in the perihematoma regions was 172.5 x 10(-5) mm(2)/s (range, 120.1-302.5 x 10(-5) mm(2)/s) and in the contralateral corresponding regions of interest was 87.6 x 10(-5) mm(2)/s (range, 76.5-102.1 x 10(-5) mm(2)/s) (P=.02). The Pearson correlation coefficient for the ADC in surrounding edema and hematoma volume was 0.7 (P=.04). The correlation coefficient between hematoma volume and contralateral hemisphere ADC was 0.8 (P=.02). CONCLUSION: We report a significant direct correlation between ICH volume and degree of ADC elevation in perihematoma and ADC values in contralateral corresponding brain tissue. These findings suggest a dose-effect interaction between volume and concentration of blood products and intensity of response that brain tissue exhibits in blood-mediated edema. Prospective natural history and interventional studies are required to confirm this biologically meaningful correlation in patients with ICH.     

Three-dimensional model of lesion geometry for evaluation of MR-guided thermal ablation therapy

RATIONALE AND OBJECTIVES: High-radiofrequency energy is used clinically to ablate pathologic tissue with interventional magnetic resonance (MR) imaging. For many tissues, resulting lesions have a characteristic appearance on contrast-enhanced T1- and T2-weighted MR images, with two boundaries enclosing an inner hypointense region and an outer hyperintense margin. Geometric modeling of three-dimensional thermal lesions in animal experiments and patient treatments would improve analyses and visualization. MATERIALS AND METHODS: The authors created a model with two quadric surfaces and 12 parameters to describe both lesion surfaces. Parameters were estimated with iterative optimization to minimize the sum of the squared shortest distances from segmented points to the model surface. The authors validated the estimation process with digital lesion phantoms that simulated varying levels of segmentation error and missing surface information. They also applied their method to in vivo images of lesions in a rabbit model. RESULTS: For simulated phantom lesions, the lesion geometry was accurate despite manual segmentation error and incomplete surface data. Even when 50% of the surface was missing, the median error was less than 0.5 mm. For all in vivo lesions, the median distance from the model surface to data was no more than 0.58 mm for both inner and outer surfaces, less than a voxel width (0.7 mm). The interquartile range was 0.89 mm or less for all data. CONCLUSION: The authors' model provides a good approximation of actual lesion geometry and is highly resistant to missing segmentation information. It should prove useful for three-dimensional lesion visualization, volume estimation, automated segmentation, and volume registration.     

Radiofrequency ablation for the treatment of liver tumors in the caudate lobe

PURPOSE: To evaluate the effectiveness of radiofrequency (RF) ablation for liver tumors located in the caudate lobe. MATERIALS AND METHODS: Ten patients (46-79 years of age; median, 70 y), eight with hepatocellular carcinoma (HCC) and cirrhosis and two with colorectal metastases in the caudate lobe, were treated with 5.8% NaCl tissue-perfused monopolar (n=7) or bipolar (n=3) RF ablation. The median tumor diameter was 41 mm (range, 25-70 mm). Procedures were performed under ultrasound and computed tomography (CT) guidance in eight and two patients, respectively. One month later, the treatment response was assessed by CT. RESULTS: Transhepatic right intercostal and transomental anterior epigastric routes were used for tumor puncture in eight and two patients, respectively. The entire RF ablation treatment required one or two procedures (median, 1.5), including two to 15 electrode repositionings (median, 6). After RF ablation procedure, one patient experienced jaundice that resolved spontaneously. In one patient, CT follow-up showed asymptomatic segmental biliary duct dilations. Median total hospital stay was 3 days (range, 2-9 d). Complete ablation was achieved in nine of 10 tumors. In one patient, ethanol ablation was necessary to complete RF ablation treatment. After a median follow up of 9.5 months (range, 5-25 mo), three patients remained tumor-free and seven had tumor relapse: two local, four distant, and one mixed. Repeat RF ablation was successfully performed in four cases. CONCLUSION: RF ablation of liver tumors located in the caudate lobe is effective despite the deep location of tumors and the vicinity of major vessels.     

Perfusion and diffusion MR imaging in enhancing malignant cerebral tumors

OBJECTIVE: Common contrast-enhancing malignant tumors of the brain are glioblastoma multiforme (GBMs), anaplastic astrocytomas (AAs), metastases, and lymphomas, all of which have sometimes similar conventional MRI findings. Our aim was to evaluate the role of perfusion MR imaging (PWI) and diffusion-weighted imaging (DWI) in the differentiation of these contrast-enhancing malignant cerebral tumors. MATERIALS AND METHODS: Forty-eight patients with contrast-enhancing and histologically proven brain tumors, 14 AAs, 17 GBMs, nine metastases, and eight lymphomas, were included in the study. All patients have undergone routine MR examination where DWI and PWI were performed in the same session. DWI was performed with b values of 0, 500, and 1000 mm(2)/s. Minimum ADC values (ADC(min)) of each tumor was later calculated from ADC map images. PWI was applied using dynamic susceptibility contrast technique and maximum relative cerebral blood volume (rCBV(max)) was calculated from each tumor, given in ratio with contralateral normal white matter. Comparisons of ADC(min) and rCBV(max) values with the histological types of the enhancing tumors were made with a one-way analysis of variance and Bonferroni test. A P value less than 0.05 indicated a statistically significant difference. RESULTS: The ADC(min) values (mean+/-S.D.) in GBMs, AAs, lymphomas, and metastases were 0.79+/-0.21 (x10(-3)mm(2)/s), 0.75+/-0.21 (x10(-3)mm(2)/s), 0.51+/-0.09 (x10(-3)mm(2)/s), and 0.68+/-0.11 (x10(-3)mm(2)/s), respectively. The difference in ADC(min) values were statistically significant between lymphomas and GBMs (P<0.05). It was also statistically significant between lymphomas and AAs (P<0.03). However, there were no differences between lymphomas and metastasis, and between GBMs, AAs, and metastasis. The rCBV(max) ratio (mean+/-S.D.) in GBMs were 6.33+/-2.03, whereas it was 3.66+/-1.79 in AAs, 2.33+/-0.68 in lymphomas, and 4.45+/-1.87 in metastases. These values were statistically different between GBMs and AAs (P<0.001), GBMs and lymphoma (P<0.0001). Although there seemed to be difference between GBMs and metastases, it was not statistically significant (P<0.083). CONCLUSION: Combination of DWI and PWI, with ADC(min) and rCBV(max) calculations, may aid routine MR imaging in the differentiation of common cerebral contrast-enhancing malignant tumors.     

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.     

Dynamic 3D-CT angiography

BACKGROUND AND PURPOSE: 3D-CT angiography (3D-CTA), a vessel-cast technique, is useful in the diagnosis of patients with brain lesions and cerebrovascular disease. However, it provides no information on blood-flow dynamics. Therefore, we developed a new technique, dynamic 3D-CTA (d3D-CTA), that yields 3D images of the vasculature and allows assessment of the hemodynamics and of the cerebral perfusion. METHODS: We subjected 12 patients to d3D-CTA; 7 had brain tumors, 4 had arteriovenous malformations (AVMs), and 1 presented with cervical internal carotid artery occlusion. d3D-CTA was performed with a multidetector-row CT scanner with 64 detectors; the parameters were 0.5-mm section thickness, 0.5 seconds per rotation, 135 kV, and 150 mA. Contrast medium (total volume, 30-35 mL) was injected at 6-7 mL/s; scanning was started after a 5-second delay. Scans were generated with volume-rendering reconstruction and displayed in cine mode. RESULTS: We succeeded in obtaining d3D-CTA scans in all cases. The scans yielded 3D information on the vascular structures and hemodynamics and on the status of the cerebral perfusion. CONCLUSION: d3D-CTA facilitates the acquisition of information on the vascular and cerebral dynamic blood flow on 3D-CTA images. Despite the relatively high radiation exposure and contrast injection speed and its limited scan range, this technique is useful for the diagnosis of patients with brain tumors or cerebrovascular disease and for treatment planning.     

Radiation Therapy for Chordoma and Chondrosarcoma of the Skull Base and the Cervical Spine

BACKGROUND: Prospective analysis of local tumor control, survival and treatment complications in 67 consecutive patients treated with fractionated photon and proton radiation for chordoma or chondrosarcoma of the base of the skull and the cervical spine. PATIENTS AND METHODS: Between December 1995 and January 2000, 67 patients with a median age of 52 years (range: 14-85 years), were treated at the Centre de Protonthérapie d'Orsay (CPO), France, using the 201-MeV proton beam, 49 for chordoma and 18 for chondrosarcoma. Irradiation combined high-energy photons and protons. Photons represented two thirds of the total dose and protons one third. The median total dose delivered within gross tumor volume (GTV) was 67 Cobalt Gray Equivalents (CGE; range: 60-70 CGE). RESULTS: Within a median follow-up of 29 months (range: 4-71 months), the 3-year local control rates were 71% and 85% for chordomas and chondrosarcomas, respectively, and the 3-year overall survival rates 88% and 75%, respectively. 14 tumors (21.5%) failed locally (eight within the GTV, four within the clinical target volume [CTV], and two without further assessment). Seven patients died from their tumor and another one from a nonrelated condition (pulmonary embolism). The maximum tumor diameter and, similarly, the GTV were larger in relapsing patients, compared with the rest of the population: 56 mm vs 44 mm (p = 0.024) and 50 ml vs 22 ml (p = 0.0083), respectively. In univariate analysis, age < or = 52 years at the time of radiotherapy (p = 0.002), maximum diameter < 45 mm (p = 0.02), and GTV < 28 ml (p = 0.02) impacted positively on local control. On multivariate analysis, only age was an independent prognostic factor of local control. CONCLUSION: In chordomas and chondrosarcomas of the skull base and cervical spine, combined photon and proton radiation therapy offers excellent chances of cure. In two thirds of the cases, relapses are located in the GTV. Maximum diameter, GTV, and age are prognostic indicators of local control. These results should be confirmed during a longer follow-up.     

Coronary calcification at electron-beam CT: effect of section thickness on calcium scoring in vitro and in vivo

PURPOSE: To compare the accuracy of electron-beam computed tomography (CT) with 3.0- and 1.5-mm section thickness for calcium quantification and the prevalence of coronary calcifications with each. MATERIALS AND METHODS: Electron-beam CT images were acquired with nonoverlapping 1.5- and 3.0-mm section thickness. Scans were obtained in an anthropomorphic thorax phantom with calcium cylinders of different sizes and densities, as well as in 1,302 study participants. A calcified lesion was defined as a minimum of 2 pixels (area, 0.52 mm2) with a minimum attenuation of 130 HU. The calcified lesions were quantified by means of a volumetric method with isotropic interpolation. From the phantom scans, mean volume scores, SDs, and measurement variations were calculated. From the participant scans, median volume scores and interquartile ranges were calculated. Participants were classified in categories based on cutoff levels for volume score quartiles for the 1.5-mm scans. An intraclass correlation coefficient (kappa value) was calculated as a measure of correlation between categories. RESULTS: In the phantom, deviations of calculated volumes from the true cylinder volumes and measurement variations were generally higher for the 3.0-mm protocol than for the 1.5-mm protocol. In the participants, the median volume score was 100 mm3 (interquartile range, 11-409 mm3) for the 3.0-mm protocol and 144 mm3 (interquartile range, 35-513 mm3) for the 1.5-mm protocol. Agreement between classifications of volume scores for the 1.5- and 3.0-mm scans was good (kappa = 0.62, P <.001). Compared with the quartile classification for the 1.5-mm scan, however, classifications for 370 (28%) participants were put in a different category with the 3.0-mm protocol. CONCLUSION: In a phantom, electron-beam CT scans with 3.0-mm section thickness yield less accurate estimates of calcified volume than do 1.5-mm scans. Electron-beam CT protocols with thinner sections considerably affect classification of individuals on the basis of the amount of coronary calcification depicted.