Validation of three-dimensional reconstructions of knee anatomy: CT vs MR imaging

This is the first investigation to validate the accuracy of spatial measurements based on computer-generated three-dimensional (3D) reconstructions of CT and magnetic resonance (MR) scans. Standard 3D reconstructions of one fresh frozen cadaveric knee specimen with implanted stereotactic reference markers were created. These reconstructions were based on the data from CT and MR scans performed before and after the resection of extraosseous soft tissues. The distances between the stereotactic reference markers in the 3D reconstructions were compared with the same measurements of the specimen using a Bridgeport milling machine as a precision 3D digitizer. The accuracy of intermarker distance measurements averaged 99% for CT-based reconstructions compared with 97.5% for MR-based reconstructions. Planar slices at a predetermined level and orientation through the femoral condyles were created by sequential milling of the specimen and were compared with reconstructed oblique planar CT and MR slices at the same level and orientation. The accuracy of condylar dimensional measurements was 94% for CT and 93% for MR compared with measurements of the milled slices. Three-dimensional reconstructions based on CT and MR scan data were of equally high quality and were unaffected by the status of the soft tissues.     

Three-dimensional MR imaging and display of intracranial disease: Improvements with the MP-RAGE sequence and gadolinium

Three-dimensional (3D) image rendering was performed in 14 patients who had undergone magnetic resonance (MR) imaging for focal brain lesions. The MR study included the magnetization-prepared rapid gradient-echo (MP-RAGE) sequence with 64 or 128 partitions. Resultant contiguous sections 2.5 or 1.25 mm thick, respectively, were obtained. Images were acquired before and after administration of gadopen-letate dimeglumine. Resultant 3D data sets were processed on a commercially available workstation. Correlative surgical observation was performed in four cases. All data sets were successfully processed into 3D images. The precontrast images proved superior to gadolinium-enhanced images for brain surface rendering. Postcontrast images proved superior for reconstruction of tumors and vascular structures. The 64-partition data set proved sufficient for all postprocessing. Coronal orientation was preferred to sagittal orientation for surface rendering because it provided optimal orthogonal orientation of sulcal and gyral brain surface features. Three-dimensional rendition allowed easy superposition of lesion, brain, vessels, and scalp features-all useful for surgical planning. The central sulcus was easily recognized in the mid-line partitions and traced mediolaterally for projection on the cortical surface. MP-RAGE provides a 3D data set that can be obtained in just over 3 minutes, from which clinically useful 3D renderings are possible. The rapidity of acquisition and capability for 3D rendering provides additional clinical utility.     

Quantitative analysis of spiral computed tomography for craniofacial clinical applications

OBJECTIVE: To determine the relative accuracy of craniofacial measurements obtained from volumetric spiral computed tomography (CT) in two-dimensional (2D) and three-dimensional (3D) reconstructions. MATERIALS AND METHODS: Nine cadaver heads were examined with spiral CT at 3 mm collimation with 2 mm/s table feed. The CT data, stored on optical disks to allow full retrospective review of any image, were transferred to a networked computer workstation for off-line visualization, measurement and analysis. Three-dimensional surface images (3DCT) were generated from spiral CT scans (2DCT). Measurements (n = 28) were made from (both unique and conventional craniometric anatomical) landmarks (n = 23) identified in 2DCT and 3DCT images by two observers twice. The soft tissues were then removed from the cadavers and the measurements performed with an electromagnetic digitizer. RESULTS: The accuracy for 3DCT was higher than 2DCT images for seven of 28 measurements. Twenty five per cent of the 2DCT measurements were found to be significantly different from the physical measurements. All measurements with 3DCT were found to have satisfactory accuracy in comparison with physical measurements. CONCLUSIONS: Three-dimensional spiral CT can distinguish craniofacial anatomy more accurately with some measurements.     

Quantification of knee joint fluid volume by MR imaging and CT using three-dimensional data processing

A noninvasive, quantitative technique to estimate joint fluid volume using three-dimensional (3D) processing of magnetic resonance (MR) imaging and CT data was evaluated. The mean accuracy error of this 3D approach with MR imaging, performed on five fresh cadaver knees, was -2.4 +/- 5.1% SD when volume estimates were based on heavily T2-weighted transverse images. A considerably higher mean accuracy error of -12.5 +/- 16.9% SD was found when the 3D approach with CT was used (four fresh cadaver knees), probably because of the small attenuation differences between articular soft-tissue structures and joint fluid. A mean precision error of 4.9 +/- 2.3% SD was found when two radiologists independently evaluated in vivo MR imaging studies of 12 knees with joint effusion. Because of the low CT accuracy, no in vivo studies were performed using CT. Thus, this preliminary study shows that quantification of joint fluid volume with 3D data processing offers more accuracy with MR imaging than with CT. The 3D approach with MR imaging provides a potential tool for clinical studies.     

Variation of perisylvian and calcarine anatomic landmarks within stereotaxic proportional coordinates

This investigation describes the variability in location of functionally important persylvian landmarks and of the calcarine sulcus within the Talairach stereotaxic grid, a system frequently used for cortical localization in functional images. Twenty healthy volunteers (40 hemispheres) had MR imaging under stereotaxic conditions. Outlines of the following structures were directly identified on sagittal 5-mm MR sections and marked on individual proportional grid overlays: inferior central sulcus, inferior precentral sulcus, inferior postcentral sulcus, anterior ascending ramus and posterior rami of the sylvian fissure, superior temporal sulcus, and calcarine sulcus. Maximal variation zones for these landmarks were defined by superimposition of the standardized individual data on a standard stereotaxic grid. The sulcal variation zones measured 1.5-2.0 cm. The findings indicate that macroanatomic individuality in the cerebral surface cannot be accounted for adequately by proportional coordinates, and that this method does not allow precise definition of anatomically based regions of interest for functional imaging. Instead, MR mapping of the individual sulcus pattern should be used to generate brain templates.     

3D volume rendering using multislice CT for dental implants

OBJECTIVES: To determine the precision and accuracy of three-dimensional (3D) volume rendering spiral multislice computed tomography (CT)-based linear measurements of the mental foramen for dental implants, in vitro, and their precision, in vivo. METHODS: Five cadaver heads were imaged by multislice spiral CT (Toshiba Aquilion) with 0.5 mm thick axial slices (0.5 mm/0.5 s of table feed) at 0.5 mm interval reconstructions. The image data sets were transferred to a networked computer workstation. Using computer graphics the data were analysed with a 3D volume rendering technique using Vitrea software. Two oral and maxillofacial radiologists, independently, made electronic linear measurements from the superior border of the mental foramen to the crest of the alveolar process. The soft tissues were removed and physical measurements made using a 3 Space (Polhemus, Colchester, VT, USA) electromagnetic digitizer with a personal computer running Windows 98. The same linear measurements of 15 patients using the same imaging methodology were performed and the precision was analysed. RESULTS: The findings showed no statistically significant inter- or intra-observer differences in vitro and in vivo, or between imaging and physical measurements in vitro (P>0.05). CONCLUSIONS: 3D multislice spiral CT imaging allows highly accurate measurements for dental implant placement in proximity to the mental foramen. Computer graphics software, using volume rendering is suitable for implant planning.     

Tissue segmentation of the brain in Alzheimer disease.

PURPOSETo compare brain tissue in patients with Alzheimer disease with that in elderly control subjects by using high-resolution MR imaging and quantitative tissue-segmentation techniques.METHODSMR imaging of the brain was performed in 21 patients with Alzheimer disease and 17 control subjects. A computerized segmentation program was used to quantify volumes of ventricular and sulcal cerebrospinal fluid (CSF), white matter, cortical gray matter, and white matter signal hyperintensity. Statistical analysis was performed using analysis of variance.RESULTSWe found a significant decrease in total brain tissue and cortical gray matter and an increase in the ventricular and sulcal CSF in Alzheimer patients compared with control subjects. There was no difference in the volume of white matter. More white matter signal hyperintensities were found in Alzheimer patients, and a significant interaction between age and group was noted. Neuropsychological test scores correlated significantly with sulcal CSF in patients with Alzheimer disease.CONCLUSIONSemiautomated segmentation of MR images of the brains of patients with Alzheimer disease reveals significant brain atrophy attributable to loss of cortical gray matter, which is compatible with the pathologic features of Alzheimer disease. There is also a significant increase in white matter signal hyperintensities. Tissue segmentation may increase our understanding of dementia but, as yet, when used alone, it does not play a role in the premorbid diagnosis of Alzheimer disease.     

Craniofacial measurements based on 3D-CT volume rendering: implications for clinical applications

OBJECTIVES: This study was designed to determine the precision and accuracy of anthropometric measurements using three-dimensional computed tomography (3D-CT) volume rendering by computer systems for craniofacial clinical applications, and to compare the craniometric landmarks using bone and soft tissue protocols. METHODS: The study population consisted of 13 cadaver heads that were examined with spiral CT. The archived CT data were transferred to a workstation, and 3D-CT volume rendered images were generated using computer graphics tools. Linear measurements (n = 10), based upon conventional craniometric anatomical landmarks (n = 08), were identified in 2D-CT and in 3D-CT images by two radiologists twice each independently, and then performed by 3D-CT imaging using a computer graphics systems using bone and soft tissue protocols. In total, 520 imaging measurements were made. The soft tissues were subsequently removed from the cadaver heads and the measurements were repeated using an electromagnetic 3 Space trade mark digitizer. RESULTS: The results demonstrated no statistically significant difference between interobserver and intraobserver measurements or between imaging and physical measurements in both 3D-CT protocols. The standard error was found to be between 0.45% and 1.44% for all the measurements in both protocols, indicating a high level of precision. Furthermore, there was no statistically significant difference between imaging and physical measurements (P > 0.01). The error between the mean actual and mean 3D-based linear measurements was 0.83% for bone and 1.78% for soft tissue measurements, demonstrating high accuracy of both 3D-CT protocols. CONCLUSIONS: 3D-CT volume rendering images using craniometric measurements can be used for anthropological studies involving craniofacial applications.     

Temporal lobe volume measurement from MR images: accuracy and left-right asymmetry in normal persons

The effect of several magnetic resonance (MR) variables on the accuracy of volume measurements in phantom objects was investigated by use of an off-line automatic border-outlining and internal area pixel-counting computer program, and an optimal set of imaging variables was identified. Measurements were made of the temporal lobe volumes of a gross fixed brain specimen from MR image data. The range in accuracy was from -2 to +7%, and the standard deviation of the difference in right minus left lobe volume measurements obtained from the MR images and those obtained by use of Archimedes' principle was 1 cm3. This volumetric technique was applied to 25 normal persons, most of whom were right-handed. The median ratio of right to left temporal lobe volume was 1.16 (range 0.99-1.23). The nondominant temporal lobe was significantly larger than the dominant. The mean difference (95% confidence interval) between right and left volumes was 7 cm3 (6-9 cm3). This confidence interval was similar to that obtained when the variability within a subject (estimated from the gross fixed brain specimen) was taken into account. Unilateral temporal lobe atrophy, particularly in patients with temporal lobe epilepsy, should be interpreted from MR images with this range of discrepancy in normal left-right size in mind.     

Retrograde cortical and deep venous drainage in patients with intracranial dural arteriovenous fistulas: comparison of MR imaging and angiographic findings

BACKGROUND AND PURPOSE: We assessed MR imaging, specifically contrast-enhanced three-dimensional (3D) magnetization-prepared rapid gradient-echo (MP-RAGE), in evaluating retrograde venous drainage in patients with intracranial dural arteriovenous fistulas (dAVFs) that may result in catastrophic venous infarction or hemorrhage. METHODS: Twenty-one patients with angiographically proved dAVFs underwent nonenhanced spin-echo (SE) and fast SE imaging, 3D fast imaging with steady-state precession, and enhanced SE and 3D MP-RAGE imaging. Retrograde venous drainage was categorized as cerebral cortical, deep cerebral, posterior fossa medullary, ophthalmic, or spinal venous. We assessed retrograde venous drainage and graded its severity. MR imaging and angiographic severities were correlated. Sensitivity, specificity, and accuracy were calculated to evaluate the diagnostic utility of each technique compared with conventional angiography. We retrospectively correlated angiograms and MR images. RESULTS: Enhanced 3D MP-RAGE and T1-weighted SE images had higher diagnostic accuracy higher than nonenhanced images, especially when retrograde drainage involved cerebral cortical, posterior fossa, and spinal veins. Correlation of severity for enhanced MP-RAGE images and enhanced T1-weighted images with angiograms was good to excellent and better than that with nonenhanced images. All sequences had low diagnostic accuracy when drainage was via deep cerebral veins. On retrospective review, 3D MP-RAGE images showed two thrombotic inferior petrosal sinuses. CONCLUSION: Enhanced MR images were superior to nonenhanced images in assessing retrograde venous drainage in intracranial dAVFs. Enhanced 3D MP-RAGE is superior to enhanced T1-weighted SE imaging for determining the route and severity of venous reflux because of its increased spatial resolution and ability to contiguously delineate the venous system.     

Cortical intraoperative stimulation in brain tumors as a tool to evaluate spatial data from motor functional MRI

RATIONALE AND OBJECTIVE: The purpose of this prospective, double-blind study was to correlate motor functional MRI (fMRI) with cortical brain mapping by intraoperative stimulation using 3D reconstructed images of the surface of the brain, and to validate the spatial data of fMRI in patients with brain tumors. METHODS: Fourteen patients with tumors of the rolandic region underwent functional MR mapping of the hand region and subsequently cortical mapping before tumor resection. Data obtained with fMRI and brain mapping were not known previously by the neurosurgeon and by the neuroradiologist, respectively (double-blind study). RESULTS: In each case, the results of direct cortical mapping matched those obtained with fMRI, both positively and negatively, although the extent of the functional activations was larger than the area required to elicit the corresponding movement during intraoperative brain mapping. CONCLUSION: fMRI can be used before surgery to assess motor functional area in patients with rolandic tumors. More studies are needed to validate during surgery the real extent of fMRI activations.     

Prostate cancer tumor volume: measurement with endorectal MR and MR spectroscopic imaging

PURPOSE: To determine accuracy of magnetic resonance (MR) and three-dimensional (3D) MR spectroscopic imaging in prostate cancer tumor volume measurement. MATERIALS AND METHODS: Endorectal MR and 3D MR spectroscopic imaging were performed in 37 patients before radical prostatectomy. Two independent readers recorded peripheral zone tumor nodule location and volume. Results were analyzed with step-section histopathologic tumor localization and volume measurement as the standard. Accuracy of tumor volume measurement was assessed with the Pearson correlation coefficient. P values were calculated with a random effects model. Bland-Altman regression analysis was used to evaluate systematic bias between tumor volumes measured with MR imaging and true tumor volumes. Analyses were performed for all nodules and nodules greater than 0.50 cm(3). RESULTS: Mean volume of peripheral zone tumor nodules (n = 51) was 0.79 cm(3) (range, 0.02-3.70 cm(3)). Two readers detected 20 (65%) and 23 (74%) of 31 peripheral zone tumor nodules greater than 0.50 cm(3). For these nodules, measurements of tumor volume with MR imaging, 3D MR spectroscopic imaging, and a combination of both were all positively correlated with histopathologic volume (Pearson correlation coefficients of 0.49, 0.59, and 0.55, respectively); only measurements with 3D MR spectroscopic imaging and a combination of MR and 3D MR spectroscopic imaging demonstrated statistical significance (P <.05). Tumor volume estimation with all three methods was more accurate for higher tumor volumes. CONCLUSION: Addition of 3D MR spectroscopic imaging to MR imaging increases overall accuracy of prostate cancer tumor volume measurement, although measurement variability limits consistent quantitative tumor volume estimation, particularly for small tumors.     

Cortical lesions in multiple sclerosis: combined postmortem MR imaging and histopathology

BACKGROUND AND PURPOSE: Cortical lesions constitute a substantial part of the total lesion load in multiple sclerosis (MS) brain. They have been related to neuropsychological deficits, epilepsy, and depression. However, the proportion of purely cortical lesions visible on MR images is unknown. The aim of this study was to determine the proportion of intracortical and mixed gray matter (GM)-white matter (WM) lesions that can be visualized with postmortem MR imaging. METHODS: We studied 49 brain samples from nine cases of chronic MS. Tissue sections were matched to dual-echo T2-weighted spin-echo (T2SE) MR images. MS lesions were identified by means of myelin basic protein immunostaining, and lesions were classified as intracortical, mixed GM-WM, deep GM, or WM. Investigators blinded to the histopathologic results scored postmortem T2SE and 3D fluid-attenuated inversion recovery (FLAIR) images. RESULTS: Immunohistochemistry confirmed 70 WM, eight deep GM, 27 mixed GM-WM, and 63 purely cortical lesions. T2SE images depicted only 3% of the intracortical lesions, and 3D FLAIR imaging showed 5%. Mixed GM-WM lesions were most frequently detectable on T2SE and 3D FLAIR images (22% and 41%, respectively). T2SE imaging showed 13% of deep GM lesions versus 38% on 3D FLAIR. T2SE images depicted 63% of the WM lesions, whereas 3D FLAIR images depicted 71%. Even after side-by-side review of the MR imaging and histopathologic results, many of the intracortical lesions could not be identified retrospectively. CONCLUSION: In contrast to WM lesions and mixed GM-WM lesions, intracortical lesions remain largely undetected with current MR imaging resolution.     

Accurate linear measurements in the anterior maxilla using orthoradially reformatted spiral computed tomography

OBJECTIVES: To validate measurement accuracy in proximity to the incisive canal from two-dimensional (2D) reformatted spiral computed tomography (CT). MATERIALS AND METHODS: Eight cadaver heads were examined with spiral CT with 1 mm thick axial slices and 1 mm.s-1 table feed. The data was transferred to a networked computer workstation to generate 2D orthoradially reformatted images. The length of the incisive canal and width of the alveolar crest were measured independently by two oral and maxillofacial radiologists. The soft tissues in the region of the incisive foramen were removed and physical measurements made using an electromagnetic digitizer. RESULTS: There was no statistically significant difference between measurements on orthoradial 2DCT images and physical measurements (P > 0.05). CONCLUSIONS: Accurate measurements for dental implant placement in proximity to the incisive canal can be made from reformatted spiral CT images.     

Differential cortical thickness across the central sulcus: a method for identifying the central sulcus in the presence of mass effect and vasogenic edema

BACKGROUND AND PURPOSE: Identification of the motor strip on MR imaging studies is difficult in the presence of mass effect and vasogenic edema because sulcal landmarks are obscured. We hypothesize that a difference in cortical thickness between the motor and sensory strips is readily apparent on T2-weighted images in the presence of vasogenic edema and reliably identifies the central sulcus. METHODS: Thirteen patients with brain tumors resulting in vasogenic edema near the central sulcus were identified. The cortical thickness of the anterior and posterior banks of the central sulcus as well as the neighboring sulci in the frontal and parietal lobes were measured from T2-weighted images. Similar measures were obtained from neighboring sulci in the frontal and parietal lobes. Location of the central sulcus was confirmed with standard anatomic landmarks in all patients and by intraoperative cortical mapping in 2 patients. RESULTS: A twofold difference in cortical thickness between the anterior and posterior banks of the central sulcus uniquely identified the central sulcus on T2-weighted images in the presence of vasogenic edema, despite the marked distortion of sulcal anatomy as a result of mass effect. This relationship was not present in neighboring sulci. CONCLUSION: Cytoarchitectonic differences in the motor and sensory cortices result in a markedly thicker posterior than anterior bank of the central sulcus that is readily visible on routine T2-weighted images in the presence of vasogenic edema. Therefore, the cortical thickness can serve as a complementary method in identification of the motor strip in patients with mass effect.     

Measurement of the volume of oral tumors by three-dimensional spiral computed tomography

OBJECTIVE: To determine the precision and accuracy of in vitro measurements of the volume of oral tumors with three-dimensional (3D) spiral computed tomography (CT) and their precision in vivo. METHODS: Two simulated tumors made of modelling compound mixed with contrast medium were positioned medial to the mandibles of five cadaver heads and examined with subsecond spiral CT. Two observers delineated the simulated tumors twice in axial, coronal and sagittal views and then measured the volume from multiplanar reconstructed images. The software tools automatically displayed the simulated tumors in 3D-reconstructed images with the volumetric measurements. The simulated tumors were removed and their volume measured by water displacement. The volume of 15 oral tumors associated with the mandible were measured in vivo with the same imaging methods and the precision analysed. RESULTS: There were no statistically significant differences between or within observers or between imaging and physical measurements in vitro, nor between inter- and intra-observer measurements in vivo (P > 0.05). CONCLUSION: Volumetric measurements from 3D-reconstructed CT are reliable and accurate in vitro and reliable in vivo. The method is potentially useful for the management of oral neoplasms.     

Cerebral cortical lesions in multiple sclerosis detected by MR imaging at 8 Tesla

BACKGROUND AND PURPOSE: Conventional imaging of ex-vivo brain at 1.5T in multiple sclerosis (MS) detects only a small fraction of the gray matter cerebral cortical lesions that can be detected by pathology. Our purpose was to examine if imaging at 8T can detect plaques in cortical gray matter (CGM) not evident at 1.5T. METHODS: An ex-vivo brain obtained at autopsy from a patient with MS was formalin fixed and 1 cm coronal slices were examined using MR imaging at 8T. RESULTS: Numerous cerebral cortical lesions not evident at 1.5T were seen at 8T. Lesions were easily identified using gradient-echo and spin-echo (SE) as well as diffusion images. MR imaging at 8T identified many of the types of plaques previously evident only by pathology. The magnitude of the cortical involvement in this 1 patient was severe. Lesions in the gray matter readily visible by high-field MR imaging were sometimes barely visible by pathology. MR imaging at 8T often facilitated the detection of such plaques by pathology. CONCLUSION: This study establishes the utility of high-field imaging at 8T in the delineation of plaques in the cerebral CGM in MS.     

Usefulness of three-dimensional MR images of brain tumors for surgical simulation

The purpose of this study was to determine the usefulness of three-dimensional (3D) MR imaging of brain tumors for surgical planning. Sixty-nine patients with various tumors of the brain were included in the present study. Using a volume-rendering (VR) method on an independent workstation, 3D-MR images were obtained with the fast-SPGR sequence after Gd-DTPA administration. VR images could show an exact relationship between the surface of the brain and major vessels. However, in patients with deeply located tumors, VR images did not necessarily provide sufficient information as to the relationship between the tumor and vessels. In combination with a surface-rendering method, 3D-MR imaging could demonstrate the exact relationships among the tumors, major vessels, and surface of the brain. In tumors without contrast enhancement, this method was able to show 3D images of tumors with surrounding structures. For neurosurgeons, 3D-MR images were useful for understanding the surface anatomy and surrounding structures of the tumors prior to surgery. These images were also helpful in explaining the condition of the disease to patients and their families.     

基于Amira软件的小腿外侧供区皮瓣三维可视化设计

目的 利用个人电脑(PC),结合图像处理软件Amira 4.1对小腿外侧皮瓣进行三维重建,建立皮瓣动脉的可视化数字模型.方法 明胶-氧化铅混悬液灌注的新鲜成人标本1具,行下肢多排螺旋CT扫描,观测腓动脉的分布及彼此间的吻合情况,将数据输入个人PC,利用Amira软件,经剪切、分割、表面重建及体绘制等步骤,根据解剖结构特点,对皮肤及其血管进行三维重建,获得立体形态的动脉皮瓣结构.结果 重建了皮瓣主要构成,重建的三维结构可以多彩色、透明或任意组合显示,经不同角度观察,整体显示清晰、实体感强,皮肤、动脉的相互关系一目了然.在三维表面重建的图像中可清楚地观察各解剖结构的形态,特别是皮瓣动脉的分支及其体表投影得到了很好的显示.结论 小腿外侧皮瓣三维重建对基础研究、临床试验及手术规划具有重要价值,应用Amira软件可为三维建模提供基础.     

Improvement of brain perfusion SPET using iterative reconstruction with scatter and non-uniform attenuation correction

Filtered back-projection (FBP) is generally used as the reconstruction method for single-photon emission tomography although it produces noisy images with apparent streak artefacts. It is possible to improve the image quality by using an algorithm with iterative correction steps. The iterative reconstruction technique also has an additional benefit in that computation of attenuation correction can be included in the process. A commonly used iterative method, maximum-likelihood expectation maximisation (ML-EM), can be accelerated using ordered subsets (OS-EM). We have applied to the OS-EM algorithm a Bayesian one-step late correction method utilising median root prior (MRP). Methodological comparison was performed by means of measurements obtained with a brain perfusion phantom and using patient data. The aim of this work was to quantitate the accuracy of iterative reconstruction with scatter and non-uniform attenuation corrections and post-filtering in SPET brain perfusion imaging. SPET imaging was performed using a triple-head gamma camera with fan-beam collimators. Transmission and emission scans were acquired simultaneously. The brain phantom used was a high-resolution three-dimensional anthropomorphic JB003 phantom. Patient studies were performed in ten chronic pain syndrome patients. The images were reconstructed using conventional FBP and iterative OS-EM and MRP techniques including scatter and non-uniform attenuation corrections. Iterative reconstructions were individually post-filtered. The quantitative results obtained with the brain perfusion phantom were compared with the known actual contrast ratios. The calculated difference from the true values was largest with the FBP method; iteratively reconstructed images proved closer to the reality. Similar findings were obtained in the patient studies. The plain OS-EM method improved the contrast whereas in the case of the MRP technique the improvement in contrast was not so evident with post-filtering.