- To view the ABSTRACT of each publication, clik on
- To view each publication in an EXTERNAL LINK, clik on
- To DOWNLOAD the FULL TEXT of each publication, clik on
Filter by type:
1.  MR-guided attenuation map for prostate PET-MRI: an intensity and morphologic-based segmentation approach for generating a fiveclass attenuation map in pelvic region.
M. Shirin Shandiz • H. Saligheh Rad • P. Ghafarian • M. Bakhshayesh Karam • Afshin Akbarzadeh • Mohammad Reza Ay
Journal Paper
Ann Nucl Med

 Abstract

PURPOSE: Prostate imaging is one of the major application of hybrid PET/MRI systems. Inaccurate attenuation maps (p-maps) derived by direct segmentation (SEG) in which the cortical bone is ignored and the volume of the air in cavities is underestimated is the main challenge of commercial PET/MRI systems for the quantitative analysis of the p pelvic region. The present study considered the cortical bone and air cavity along with soft tissue, fat, and background air in the p-map of the pelvic region using a method based on SEG. The proposed method uses a dedicated imaging technique that increases the contrast between regions and a hybrid segmentation method to classify MR images based on intensity and morphologic characteristics of tissues, such as symmetry and similarity of bony structures.
PROCEDURES: Ten healthy volunteers underwent MRI and ultra-low dose CT imaging. The dedicated MR imaging technique uses the short echo time (STE) based on the conventional sequencing implemented on a clinical 1.5T MRI scanner. The generation of a p-map comprises the following steps: (1) bias field correction; (2) hybrid segmentation (HSEG), including segmenting images into clusters of cortical bone-air, soft tissue, and fat using https://www.ncbi.nlm.nih.gov/pubmed/?term=MR-guided+attenuation+map+for+prostate+PET-MRI%3A+an+intensity+and+morphdogic-based+segmen... 1/2 12/7/2016 MR-guided attenuation map for prostate PET-MRI: an intensity and morphologic-based segmentation approach for generating a five-class attenuatio...
spatial fuzzy c-means (SFCM), and separation of cortical bone and internal air cavities using morphologic characteristics; (3) the active contour approach for the separation of background air; and (4) the generation of a five-class p-map for cortical bone, internal air cavity, soft tissue, fat tissue, and background air. Validation was done by comparison with segmented CT images.
RESULTS: The Dice and sensitivity metrics of cortical bone structures and internal air cavities were 72 ± 11 and 66 ± 13 and 73 ± 10 and 68 ± 20 %, respectively. High correlation was observed between CT and HSEG-based p-maps (R 2 > 0.99) and the corresponding sinograms (R 2 > 0.98).

2.  New approach for calibration of pixelated scintillation detectors of intraoperative gamma cameras
Afshin Akbarzadeh, Valiallah Saba, Mohammad Reza Ay
Journal Paper
Iranian Journal of Nuclear Medicine and Molecular Imaging, 2017, Vol. 25, No. 1, pp 34-42.

 

Abstract:

Introduction: SURGEOSIGHT is a dedicated intra-operative hand-held gamma camera designed and manufactured by our team to be used for lymphoscintigraphy. Although conventional gamma cameras are widely used, they lack the capability of the sentinel lymph node (SLN) imaging in the operation room. Like conventional gamma cameras it is necessary to calibrate SURGEOSIGHT for linearity distortion, energy and flood correction. Our main objective in this study is to develop and implement three techniques for linearity, energy and uniformity corrections.

Methods: The camera is made up of a pixelated cesium iodide (sodium-activated) (CsI(Na)) scintillation crystal with 1 × 1 mm2 element size along with a Hamamatsu H8500 flat-panel multi-anode (64 channels) photomultiplier tube. All triple corrections required lookup tables (LUTs). To generate all three LUTs, one uncalibrated acquisition is necessary. Linearity distortion correction was based on localization of physical pixels of crystal and thin plate spline interpolation to estimate the amount of distortion. Energy correction LUT was simply a photo-peak map of incoming events which was estimated using self-organizing map technique.

Results: The results showed improvements qualitatively and quantitatively. The extrinsic energy resolution was enhanced from 50% to about 20.6%. Integral uniformity and the differential uniformity, after uniformity correction, in useful field-of-view (UFOV) was measured 9.5% and 4.5%, respectively.

Conclusion: Results substantiated that the correction techniques guarantee uniform and accurate output of the SURGEOSIGHT which is desirable for intra-operative localization of the sentinel lymph nodes in breast cancer and also for imaging of other superficial tumors.

Key words: Uniformity correction; Energy correction; Linearity distortion correction; Self-organizing map; Intra-operative gamma camera

3.  Single STE-MR Acquisition in MR-Based Attenuation Correction of Brain PET Imaging Employing a Fully Automated and Reproducible Level-Set Segmentation Approach
Fathi Kazerooni A. Av MR. Arfaie S Khateri P. Saliaheh Rad H
Journal Paper
Mol Imaging Biol.

 Abstract

Purpose: The aim of this study is to introduce a fully automatic and reproducible short echo-time (STE) magnetic resonance imaging (MRI) segmentation approach for MR-based attenuation correction of positron emission tomography (PET) data in head region. 
Procedures: Single STE-MR imaging was followed by generating attenuation correction maps (μ-maps) through exploiting an automated clustering-based level-set segmentation approach to classify head images into three regions of cortical bone, air, and soft tissue. Quantitative assessment was performed by comparing the STE-derived region classes with the corresponding regions extracted from X-ray computed tomography (CT) images. Results: The proposed segmentation method returned accuracy and specificity values of over 90 % for cortical bone, air, and soft tissue regions. The MR- and CT-derived μ-maps were compared by quantitative histogram analysis.
Conclusions: The results suggest that the proposed automated segmentation approach can reliably discriminate bony structures from the proximal air and soft tissue in single STE-MR images, which is suitable for generating MR-based μ-maps for attenuation correction of PET data 

4.  Neural correlates of audiotactile phonetic processing in early‑blind readers: an fMRI study
Morteza Pishnamazi · Yasaman Nojaba · Habib Ganjgahi · Asie Amousoltani · Mohammad Ali Oghabian
Journal Paper
EXPERIMENTAL BRAIN RESEARCH

 Abstract

Reading is a multisensory function that relies on arbitrary associations between auditory speech sounds and symbols from a second modality. Studies of bimodal phonetic perception have mostly investigated the integration of visual letters and speech sounds. Blind readers perform an analogous task by using tactile Braille letters instead of visual letters. The neural underpinnings of audiotactile phonetic processing have not been studied before. We used functional magnetic resonance imaging to reveal the neural correlates of audiotactile phonetic processing in 16 early-blind Braille readers. Braille letters and corresponding speech sounds were presented in unimodal, and congruent/incongruent bimodal configurations. We also used a behavioral task to measure the speed of blind readers in  dentifying letters presented via tactile and/or auditory modalities. Reaction times for tactile stimuli were faster. The reaction times for bimodal stimuli were equal to those for the slower auditory-only stimuli. fMRI analyses revealed the convergence of unimodal auditory and unimodal tactile responses in areas of the right precentral gyrus and bilateral crus I of the cerebellum. The left and right planum temporale fulfilled the 'max criterion' for bimodal integration, but activities of these areas were not sensitive to the phonetical congruency between sounds and Braille letters. Nevertheless, congruency effects were found in regions of frontal lobe and cerebellum. Our findings suggest that, unlike sighted readers who are assumed to have amodal phonetic representations, blind readers probably process letters and sounds separately. We discuss that this distinction might be due to mal-development of multisensory neural circuits in early blinds or it might be due to inherent differences between Braille and print reading mechanisms. 

5.  T-1 correlates age: A short-TE MR relaxometry study in vivo on human cortical bone free water at 1.5 T
Atena Akbari , Shahrokh Abbasi-Rad , Hamidreza Saligheh Rad
Journal Paper
BONE

 Abstract

Large pores of human cortical bone (>30 mu m) are filled with fluids, essentially consisting of water, suggesting that cortical bone free water can be considered as a reliable surrogate measure of cortical bone porosity and hence quality. Signal from such pores can be reliably captured using Short Echo Time (STE) pulse sequence with echo-time in the range of 1-1.5 msec (which should be judiciously selected correspond to T-2* value of free water molecules). Furthermore, it is well-known that cortical bone T-1-relaxivity is a function of its geometry, suggesting that cortical bone free water increases with age. In this work, we quantified cortical bone free water longitudinal relaxation time (T-1) by a Dual-TR technique using STE pulse sequence. In the sequel, we investigated relationship between STE-derived
cortical bone free water T-1-values and age in a group of healthy volunteers (thirty subjects covering the age range of 20-70 years) at 1.5 T. Preliminary results showed that cortical bone free water T-1 highly correlates with age (r(2) = 0.73, p < 0.0001), representing cortical bone free water T-1 as a reliable indicator of cortical bone porosity and age-related deterioration. It can be concluded that STEMRI can be utilized as proper alternative in quantifying cortical bone porosity parameters in-vivo, with the advantages of widespread clinical availability and being cost-effective, (c) 2015 Elsevier Inc. All rights reserved.
 

6.  Monte Carlo-based assessment of the trade-off between spatial resolution, field-of-view and scattered radiation in the variable resolution X-ray CT scanner
Hossein Arabi , Ali Reza Kamali Asl , Mohammad Reza Ay , Habib Zaidi
Journal Paper
PHYSICA MEDICA-EUROPEAN JOURNAL OF MEDICAL PHYSICS

 Abstract

Objective: The purpose of this work is to evaluate the impact of optimization of magnification on performance parameters of the variable resolution X-ray (VRX) CT scanner. MethodsA realistic model based on an actual VRX CT scanner was implemented in the GATE Monte Carlo simulation platform. To evaluate the influence of system magnification, spatial resolution, fieldof- view (FOV) and scatter-to-primary ratio of the scanner were estimated for both fixed and optimum object magnification at each detector rotation angle. Comparison and inference between these
performance parameters were performed angle by angle to determine appropriate object position at each opening half angle.
Results: Optimization of magnification resulted in a trade-off between spatial resolution and FOV of the scanner at opening half angles of 90 degrees-12 degrees, where the spatial resolution increased up to 50% and the scatter-to-primary ratio decreased from 4.8% to 3.8% at a detector angle of about 90 degrees for the same FOV and X-ray energy spectrum. The disadvantage of magnification optimization at these angles is the significant reduction of the FOV (up to 50%). Moreover, magnification optimization was definitely beneficial for opening half angles below 12 degrees improving the spatial resolution from 7.5 cy/mm to 20 cy/mm. Meanwhile, the FOV increased by more than 50% at these angles.

7.  Monte Carlo-based assessment of the trade-off between spatial resolution, field-of-view and scattered radiation in the variable resolution X-ray CT scanner
Hossein Arabi , Ali Reza Kamali Asl , Mohammad Reza Ay , Habib Zaidi
Journal Paper
Physica Medica

 Abstract

The purpose of this work is to evaluate the impact of optimization of magnification on performance parameters of the variable resolution X-ray (VRX) CT scanner. A realistic model based on an actual VRX CT scanner was implemented in the GATE Monte Carlo simulation platform. To evaluate the influence of system magnification, spatial resolution, field-of-view (FOV) and scatter-to-primary ratio of the scanner were estimated for both fixed and optimum object magnification at each detector rotation angle. Comparison and inference between these performance parameters were performed angle by angle to determine appropriate object position at each opening half angle.
Optimization of magnification resulted in a trade-off between spatial resolution and FOV of the scanner at opening half angles of 90°-12°, where the spatial resolution increased up to 50% and the scatter-to-primary ratio decreased from 4.8% to 3.8% at a detector angle of about 90° for the same FOV and X-ray energy spectrum. The disadvantage of magnification optimization at these angles is the significant reduction of the FOV (up to 50%). Moreover, magnification optimization was definitely beneficial for opening half angles below 12° improving the spatial resolution from 7.5 cy/mm to 20 cy/mm. Meanwhile, the FOV increased by more than 50% at these angles.

8.  Development of Gold-Coated Magnetic Nanoparticles as a Potential MRI Contrast Agent
Ali Reza Montazerabadi, Mohammad Ali Oghabian*, Rasoul Irajirad, Samad Muhammadnejad, Davoud Ahmadvand, Hamid Delavari H and Seyed Rabie Mahdavi
Journal Paper
NANO

 Abstract

Gold-coated superparamagnetic iron oxide nanoparticles (SPIONs) coated with methylpolyethylene glycol (mPEG) are synthesized and investigated as a magnetic resonance (MR) imaging contrast agent. The synthesized mPEG-core@shells are characterized by UV-visible spectroscopy, transmission electron microscopy (TEM), dynamic light scattering (DLS), vibrating sample magnetometry (VSM), zeta-potential analysis and X-ray diffraction (XRD). In addition, the transverse relaxivity of the mPEG-core@shells is measured using a 3T MRI scanner. The cytotoxicity of the mPEG-core@shells is tested in the LNCaP cell line using an 3-(4,5-Dimethylthiazol-2-yl)-2,5- diphenyltetrazolium bromide (MTT) assay. The results show that the mPEG-core@shell particles are semispherical with hydrodynamic size of similar to 65 nm and a transverse relaxivity of 162.3 mM(-1) S-1. The mPEG-core@shell particles demonstrate good stability in biological media without any significant in vitro cytotoxicity under high cellular uptake conditions. Finally, in vivo imaging shows that mPEG-core@shells are a potential contrast agent for use in early-stage detection.

9.  Effect of Functional Group and Surface Charge of PEG and Dextran-Coated USPIO as a Contrast Agent in MRI on Relaxivity Constant
Nastaran Najafian • Saeed Shanehsazzadeh • Farzaneh Hajesmaeelzadeh • Afsaneh Lahooti • Cordula Gruettner • Mohammad Ali Oghabian
Journal Paper
APPLIED MAGNETIC RESONANCE

 Abstract

The challenges in molecular imaging are focused on the development of novel contrast agents with much lower relaxation times. Ultrasmall superparamagnetic iron oxide nanoparticles known as USPIOhave been used for a variety of applications such as imaging of cancer, apoptosis, and hyperthermia providing higher signal changes based on weighted susceptibility effects. Although many studies had been performed on USPIO there is still a lack of data on the effects of the physicochemical properties of these nanoparticles (NP) such as hydrodynamic size, surface charge and type of functional groups, which may alter the relaxivity of these NPs. The aim of this study was to evaluate the effect of different functional groups and surface charges of PEGylated and dextran-coated NPs on their magnetic properties. All relaxometry studies were performed using a 3-T magnetic resonance imaging. Our results showed that the impact of charge on magnetic properties is much higher than that of coating thickness. In this respect, particles with positive surface charges showed higher r (2)/r (1) ratios.

10.  Rapid microwave-assisted synthesis of PVP-coated ultrasmall gadolinium oxide nanoparticles for magnetic resonance imaging
Parisa Vahdatkhah , Hamid Reza Madaah Hosseini , Azin Khodaei , Ali Reza Montazerabadi , Rasoul Irajirad , Mohamad Ali Oghabian , Hamid Delavari H.
Journal Paper
Chemical Physics

 Abstract

Synthesis of polyvinyl pyrrolidone (PVP) coated ultrasmall Gd203 nanoparticles (NPs) with enhanced T-1-weighted signal intensity and r(2)/r(1) ratio close to unity is performed by a microwave-assisted polyol process. PVP coated Gd(2)0(3)NPs with spherical shape and uniform size of 2.5 +/- 0.5 nm have been synthesized below 5 min and structure and morphology confirmed by HRTEM, XRD and FTIR. The longitudinal (r(1)) and transversal relaxation (r(2)) of Gd(2)0(3)NPs is measured by a 3 T MRI scanner. The results showed considerable increasing of relaxivity for Gd(2)0(3)NPs in comparison to gadolinium chelates which are commonly used for clinical magnetic resonance imaging. In addition, a mechanism for Gd(2)0(3)NPs formation and in situ surface modification of PVP-grafted Gd(2)0(3)NPs is proposed. (C) 2015 Elsevier B.V. All rights reserved.                                                            

11.  Multi-parametric (ADC/PWI/T2-w) image fusion approach for accurate semi-automatic segmentation of tumorous regions in glioblastoma multiforme
Fathi Kazerooni A, Mohseni M, Rezaei S, Bakhshandehpour G, Saligheh Rad H.
Journal Paper
Magn Reson Mater Phy

 Abstract

OBJECT:

Glioblastoma multiforme (GBM) brain tumor is heterogeneous in nature, so its quantification depends on how to accurately segment different parts of the tumor, i.e. viable tumor, edema and necrosis. This procedure becomes more effective when metabolic and functional information, provided by physiological magnetic resonance (MR) imaging modalities, like diffusion-weighted-imaging (DWI) and perfusion-weighted-imaging (PWI), is incorporated with the anatomical magnetic resonance imaging (MRI). In this preliminary tumor quantification work, the idea is to characterize different regions of GBM tumors in an MRI-based semi-automatic multi-parametric approach to achieve more accurate characterization of pathogenic regions.

 

12.  Performance evaluation of a newly developed high-resolution, dual-head animal SPECT system based on the NEMA NU1-2007 standard
Moji V, Zeratkar N, Farahani MH, Aghamiri MR, Sajedi S, Teimourian B, Ghafarian P, Sarkar S, Ay MR.
Journal Paper
JOURNAL OF APPLIED CLINICAL MEDICAL PHYSICS

 Abstract

Small-animal single-photon emission computed tomography (SPECT) system plays an important role in the field of drug development and investigation of potential drugs in the preclinical phase. The small-animal High-Resolution SPECT (HiReSPECT) scanner has been recently designed and developed based on compact and high-resolution detectors. The detectors are based on a high-resolution parallel hole collimator, a cesium iodide (sodium-activated) pixelated crystal array and two H8500 position-sensitive photomultiplier tubes. In this system, a full set of data cor- rections such as energy, linearity, and uniformity, together with resolution recovery option in reconstruction algorithms, are available. In this study, we assessed the performance of the system based on NEMA-NU1-2007 standards for pixelated detector cameras. Characterization of the HiReSPECT was performed by measure- ment of the physical parameters including planar and tomographic performance. The planar performance of the system was characterized with flood-field phantom for energy resolution and uniformity. Spatial resolution and sensitivity were evaluated as functions of distance with capillary tube and cylindrical source, respectively. Tomographic spatial resolution was characterized as a function of radius of rotation (ROR). A dedicated hot rod phantom and image quality phantom was used for the evaluation of overall tomographic quality of the HiReSPECT. The results showed that the planar spatial resolution was ~ 1.6 mm and ~ 2.3 mm in terms of full-width at half-maximum (FWHM) along short- and long-axis dimensions, respectively, when the source was placed on the detector surface. The integral uniformity of the system after uniformity correction was 1.7% and 1.2% in useful field of view (UFOV) and central field of view (CFOV), respectively. System sensitivity on the collimator surface was 1.31 cps/μCi and didn't vary significantly with distance.
Mean tomographic spatial resolution was measured ~ 1.7 mm FWHM at the radius of rotation of 25 mm with dual-head configuration.The measured performance demonstrated that the HiReSPECT scanner has acceptable image quality and, hence, is well suited for preclinical molecular imaging research.
 

13.  Resolution-recovery-embedded image reconstruction for a highresolution animal SPECT system
Zeraatkar N, Sajedi S, Farahani MH, Arabi H, Sarkar S, Ghafarian P, Rahmim A, Ay MR.
Journal Paper
Physica Medica

 Abstract

The small-animal High-Resolution SPECT (HiReSPECT) is a dedicated dual-head gamma camera recently designed and developed in our laboratory for imaging of murine models. Each detector is composed of an array of 1.2 x 1.2 mm(2) (pitch) pixelated CsI(Na) crystals. Two position-sensitive photomultiplier tubes (H8500) are coupled to each head's crystal. In this paper, we report on a resolution-recovery-embedded image reconstruction code applicable to the system and present the experimental results achieved using different phantoms and mouse scans. Collimator-detector response functions (CDRFs) were measured via a pixel-driven method using capillary sources at finite distances from the head within the field of view (FOV). CDRFs were then fitted by independent Gaussian functions. Thereafter, linear interpolations were applied to the standard deviation (sigma) values of the fitted Gaussians, yielding a continuous map of CDRF at varying distances from the head.
A rotation-based maximum-likelihood expectation maximization (MLEM) method was used for reconstruction. A fast rotation algorithm was developed to rotate the image matrix according to the desired angle by means of pre-generated rotation maps. The experiments demonstrated improved resolution utilizing our resolution-recovery-embedded image reconstruction. While the full-width at halfmaximum (FWHM) radial and tangential resolution measurements of the system were over 2 mm in nearly all positions within the FOV without resolution recovery, reaching around 2.5 mm in some locations, they fell below 1.8 mm everywhere within the FOV using the resolution-recovery algorithm.
The noise performance of the system was also acceptable; the standard deviation of the average counts per voxel in the reconstructed images was 6.6% and 8.3% without and with resolution recovery, respectively. (C) 2014 Associazione Italians di Fisica Medics. Published by Elsevier Ltd. All rights reserved 

14.  Resolution-recovery-embedded image reconstruction for a highresolution animal SPECT system
Navid Zeraatkar a, Salar Sajedi , Mohammad Hossein Farahani , Hossein Arabi , Saeed Sarkar , Pardis Ghafarian , Arman Rahmim , Mohammad Reza Ay
Journal Paper
PHYSICA MEDICA-EUROPEAN JOURNAL OF MEDICAL PHYSICS

 Abstract

The small-animal High-Resolution SPECT (HiReSPECT) is a dedicated dual-head gamma camera recently designed and developed in our laboratory for imaging of murine models. Each detector is composed of an array of 1.2 x 1.2 mm(2) (pitch) pixelated CsI(Na) crystals. Two position-sensitive photomultiplier tubes (H8500) are coupled to each head's crystal. In this paper, we report on a resolution-recovery-embedded image reconstruction code applicable to the system and present the experimental results achieved using different phantoms and mouse scans. Collimator-detector response functions (CDRFs) were measured via a pixel-driven method using capillary sources at finite distances from the head within the field of view (FOV). CDRFs were then fitted by independent Gaussian functions. Thereafter, linear interpolations were applied to the standard deviation (sigma) values of the fitted Gaussians, yielding a continuous map of CDRF at varying distances from the head.
A rotation-based maximum-likelihood expectation maximization (MLEM) method was used for reconstruction. A fast rotation algorithm was developed to rotate the image matrix according to the desired angle by means of pre-generated rotation maps. The experiments demonstrated improved resolution utilizing our resolution-recovery-embedded image reconstruction. While the full-width at halfmaximum (FWHM) radial and tangential resolution measurements of the system were over 2 mm in nearly all positions within the FOV without resolution recovery, reaching around 2.5 mm in some locations, they fell below 1.8 mm everywhere within the FOV using the resolution-recovery algorithm.
The noise performance of the system was also acceptable; the standard deviation of the average counts per voxel in the reconstructed images was 6.6% and 8.3% without and with resolution recovery, respectively. (C) 2014 Associazione Italians di Fisica Medics. Published by Elsevier Ltd. All rights reserved.

15.  Breast Cancer Cells Imaging By Targeting Methionine Transporters with Gadolinium-Based Nanoprobe
Bita Mehravi, Mehdi Shafiee Ardestani, Maryam Damercheli, Haleh Soltanghoraee, Negar Ghanaldarlaki, Ali M. Alizadeh, Mohammad A. Oghabian, Maryam Shahzad Shirazi, Shabnam Mahernia, Massoud Amanlou
Journal Paper
MOLECULAR IMAGING AND BIOLOGY

 Abstract

Early cancer diagnosis using MRI imaging is of high global interest as a non-invasive and powerful modality. In this study, methionine was conjugated on gadolinium-based mesoporous silica nanospheres to evaluate intra-cellular uptake and its accumulation in human breast cancer cells. The contrast agent was synthesized and characterized using different techniques including N-2 physisorption, thermal gravimetric analysis, dynamic light scattering, and inductively coupled plasma atomic emission spectroscopy (ICP-AES). The intra-cellular uptake of Gd3+ was measured by ICPAES, fluorescent microscopy, and flow cytometry. Finally, cellular and tumor MR imaging were performed to determine in vitro and in vivo relaxometry.
According to the results, the contrast agents accumulated in tumor cells both in vitro and in vivo. There was no significant cellular toxicity on either normal or cancer cells along with strong intense signal on T (1) compared to the unlabeled cells.

16.  Derivation of attenuation map for attenuation correction of PET data in the presence of nanoparticulate contrast agents using spectral CT imaging.
Hossein Ghadiri • Mohammad Bagher Shiran • Hamid Soltanian-Zadeh • Arman Rahmim • Habib Zaidi • Mohammad Reza Ay Received:
Journal Paper
ANNALS OF NUCLEAR MEDICINE

 Abstract

OBJECTIVE:

Uptake value in quantitative PET imaging is biased due to the presence of CT contrast agents when using CT-based attenuation correction. Our aim was to examine spectral CT imaging to suppress inaccuracy of 511 keV attenuation map in the presence of multiple nanoparticulate contrast agents.

METHODS:

Using a simulation study we examined an image-based K-edge ratio method, in which two images acquired from energy windows located above and below the K-edge energy are divided by one another, to identify the exact location of all contrast agents. Multiple computerized phantom studies were conducted using a variety of NP contrast agents with different concentrations. The performance of the proposed methodology was compared to conventional single-kVp and dual-kVp methods using wide range of contrast agents with varying concentrations.

RESULTS:

The results demonstrate that both single-kVp and dual-kVp energy mapping approaches produce inaccurate attenuation maps at 511 keV in the presence of multiple simultaneous contrast agents. In contrast, the proposed method is capable of handling multiple simultaneous contrast agents, thus allowing suppression of 511 keV attenuation map inaccuracy.

CONCLUSION:

Attenuation map produced by spectral CT clearly outperforms conventional single-kVp and dual-kVp approaches in the generation of accurate attenuation maps in the presence of multiple contrast agents.

17.  Classification of bones from MR images in torso PET-MR imaging using a statistical shape model
MohammadReza Ay,Afshin Akbarzadeh,Alireza Ahmadian,Habib Zaid
Journal Paper
NuclearInstrumentsandMethodsinPhysicsResearchA734(2014)196–200

 

Abstract

There have been exclusive features for hybrid PET/MRI systems in comparison with its PET/CT counterpart in terms of reduction of radiation exposure, improved soft-tissue contrast and truly simultaneous and multi-parametric imaging capabilities. However, quantitative imaging on PET/MR is challenged by attenuation of annihilation photons through their pathway. The correction for photon attenuation requires the availability of patient-specific attenuation map, which accounts for the spatial distribution of attenuation coefficients of biological tissues. However, the lack of information on electron density in the MR signal poses an inherent difficulty to the derivation of the attenuation map from MR images. In other words, the MR signal correlates with proton densities and tissue relaxation properties, rather than with electron density and, as such, it is not directly related to attenuation coefficients. In order to derive the attenuation map from MR images at 511 keV, various strategies have been proposed and implemented on prototype and commercial PET/MR systems. Segmentation-based methods generate an attenuation map by classification of T1-weighted or high resolution Dixon MR sequences followed by assignment of predefined attenuation coefficients to various tissue types. Intensity-based segmentation approaches fail to include bones in the attenuation map since the segmentation of bones from conventional MR sequences is a difficult task. Most MR-guided attenuation correction techniques ignore bones owing to the inherent difficulties associated with bone segmentation unless specialized MR sequences such as ultra-short echo (UTE) sequence are utilized. In this work, we introduce a new technique based on statistical shape modeling to segment bones and generate a four-class attenuation map. Our segmentation approach requires a torso bone shape model based on principle component analysis (PCA). A CT-based training set including clearly segmented bones of the torso region of 20 clinical studies was designed. Using this training set, a bone atlas was trained taking advantage of PCA analysis. Our active shape segmentation technique uses the trained shape model to segment bones from user defined initial seed points. The segmentation algorithm was evaluated using 10 clinical datasets (aligned MR and CT pairs). The resulting attenuation maps were compared to corresponding attenuation maps derived from CT resulting in a mean relative difference less than 7%.

18.  Patterns of brain activation during craving in heroin dependents successfully treated by methadone maintenance and abstinence-based treatments.
Tabatabaei-Jafari H, Ekhtiari H, Ganjgahi H, Hassani-Abharian P, Oghabian MA, Moradi A, Sadighi N, Zarei M
Journal Paper
JOURNAL OF ADDICTION MEDICINE

 Abstract

Objective: Abstinence-based therapy (ABT) and methadone maintenance therapy (MMT) are common methods of treatment in heroin dependence as both suppress subjective feeling of drug craving. However, it is not clear whether the neural basis of craving suppression is similar in both types of treatments. In this study, we compared brain activation during pictorial presentation of heroin-related cues in ABT and MMT groups to understand the neural basis of drug craving in these groups.
Methods: Three groups of subjects (successful ABT and MMT clients and healthy control) underwent functional magnetic resonance imaging, while heroin-related cues and neutral cues were presented to them. In addition, subjective cue-elicited craving has been measured using drug drive questionnaire before and after imaging.
Result: Self-report of craving was not different between ABT and MMT groups before and after scanning. Anterior cingulate cortex and inferior frontal gyrus showed higher activations in ABT than in healthy control. Inferior frontal gyrus and superior temporal gyrus showed higher activity in ABT than in MMT.
Lingual gyrus and cerebellum showed higher activity in MMT than in healthy control. Conclusions: Heroin avoidance may be achieved by MMT or ABT; however, the neural mechanism underlying these therapeutic methods differs.
 

19.  Design and development of a high resolution animal SPECT scanner dedicated for rat and mouse imaging
Salar Sajedi , Navid Zeraatkar , Vahideh Moji , Mohammad Hossein Farahani , Saeed Sarkar, Hossein Arabi , Behnoosh Teymoorian , Pardis Ghafarian , Arman Rahmim , Mohammad Reza Ay
Journal Paper
Nuclear Instruments and Methods in Physics Research A

 Abstract

A dedicated small-animal SPECT system, HiReSPECT, was designed and developed to provide a high resolution molecular imaging modality in response to growing research demands. HiReSPECT is a dual-head system mounted on a rotating gantry. The detection system is based on pixelated CsI(Na) scintillator crystals coupled to two Hamamatsu H8500 Position Sensitive Photomultiplier Tubes in each head. Also, a high resolution parallel-hole collimator is applied to every head. The dimensions of each head are 50 mm x 100 mm, enabling sufficient transaxial and axial fields-of-view (TFOV and AFOV), respectively, for coverage of the entire mouse in single-bed position imaging. However, a 50 mm TFOV is not sufficient for transaxial coverage of rats. To address this, each head can be rotated by 90 degrees in order to align the larger dimension of the heads with the short body axis, allowing tomographic data acquisition for rats. An innovative non-linear recursive filter was used for signal processing/detection. Resolution recovery was also embedded in the modified Maximum-Likelihood Expectation Maximization (MLEM) image reconstruction code to compensate for Collimator-Detector Response (CDR). Moreover, an innovative interpolation algorithm was developed to speed up the reconstruction code. The planar spatial resolution at the head surface and the image spatial resolutions were 1.7 mm and 1.2-1.6 mm, respectively. The measurements followed by post-processing showed that the observed count rate at 20% count loss is about 42 kcps. The system sensitivity at the collimator surface for heads 1 and 2 were 1.32 cps/mu Ci and 1.25 cps/mu Ci, respectively. The corresponding values were 1.18 cps/mu Ci and 1.02 cps/mu Ci at 8 cm distance from the collimator surfaces. In addition, whole-body scans of mice demonstrated appropriate imaging capability of the HiReSPECT. (C) 2014 Elsevier B.V. All rights reserved.

20.  3D Prior Image Constrained Projection Completion for X-ray CT Metal Artifact Reduction
Abolfazl Mehranian, Mohammad Reza Ay, Arman Rahmim, Habib Zaidi
Journal Paper
IEEE TRANSACTIONS ON NUCLEAR SCIENCE

 Abstract

The presence of metallic implants in the body of patients undergoing X-ray computed tomography (CT) examinations often results insevere streaking artifacts that degrade image quality. In this work, we propose a new metal artifact reduction (MAR) algorithm for 2D fan-beam and 3D cone-beam CT based on the maximum a posteriori (MAP) completion of the projections corrupted by metallic implants. In this algorithm, the prior knowledge obtained from a tissue-classified prior image is exploited in the completion of missing projections and incorporated into a new prior potential function. The prior is especially designed to exploit and promote the sparsity of a residual projection (sinogram) dataset obtained from the subtraction of the unknown target dataset from the projection dataset of the tissueclassified prior image. The MAP completion is formulated as an equality-constrained convex optimization and solved using an accelerated projected gradient algorithm. The performance of the proposed algorithm is compared with two state-of-the-art algorithms, namely 3D triangulated linear interpolation (LI) and normalized metal artifact reduction (NMAR) algorithm using simulated and clinical studies. The simulations targeting artifact reduction in 2D fan-beam and 3D cone-beam CT demonstrate that our algorithm can outperform its counterparts, particularly in cone-beam CT. In the clinical datasets, the performance of the proposed algorithm was subjectively and objectively compared in terms of metal artifact reduction of a sequence of 2D CT slices. The clinical results show that the proposed algorithm effectively reduces metal artifacts without introducing new artifacts due to erroneous interpolation and normalization as in the case of LI and NMAR algorithms. 

21.  False positive control of activated voxels in single fMRI analysis using bootstrap resampling in comparison to spatial smoothing
Fahimeh Darki, Mohammad Ali Oghabian
Journal Paper
MAGNETIC RESONANCE IMAGING

 Abstract

Functional magnetic resonance imaging (fMRI) is an effective tool for the measurement of brain neuronal activities. To date, several statistical methods have been proposed for analyzing fMRI datasets to select true active voxels among all the voxels appear to be positively activated. Finding a reliable and valid activation map is very important and becomes more crucial in clinical and neurosurgical investigations of single fMRI data, especially when pre-surgical planning requires accurate lateralization index as well as a precise localization of activation map.
Defining a proper threshold to determine true activated regions, using common statistical processes, is a challenging task. This is due to a number of variation sources such as noise, artifacts, and physiological fluctuations in time series of fMRI data which affect spatial distribution of noise in an expected uniform activated region. Spatial smoothing methods are frequently used as a preprocessing step to reduce the effect of noise and artifacts. The smoothing may lead to a shift and enlargement of activation regions, and in some extend, unification of distinct regions.
In this article, we propose a bootstrap resampling technique for analyzing single fMRI dataset with the aim of finding more accurate and reliable activated regions. This method can remove false positive voxels and present high localization accuracy in activation map without any spatial smoothing and statistical threshold setting. (C) 2013 Elsevier Inc. All rights reserved.

22.  Smoothly Clipped Absolute Deviation (SCAD) regularization for compressed sensing MRI Using an augmented Lagrangian scheme
Journal Paper
MAGNETIC RESONANCE IMAGING

 Abstract

Purpose: Compressed sensing (CS) provides a promising framework for MR image reconstruction from highly undersampled data, thus reducing data acquisition time. In this context, sparsity-promoting regularization techniques exploit the prior knowledge that MR images are sparse or compressible in a given transform domain. In this work, a new regularization technique was introduced by iterative linearization of the non-convex smoothly clipped absolute deviation (SCAD) norm with the aim of reducing the sampling rate even lower than it is required by the conventional l(1) norm while approaching an l(0) norm.
Materials and Methods: The CS-MR image reconstruction was formulated as an equality-constrained optimization problem using a variable splitting technique and solved using an augmented Lagrangian (AL) method developed to accelerate the optimization of constrained problems. The performance of the resulting SCAD-based algorithm was evaluated for discrete gradients and wavelet sparsifying transforms and compared with its l(1)-based counterpart using phantom and clinical studies. The kspaces of the datasets were retrospectively undersampled using different sampling trajectories. In the AL framework, the CS-MRI problem was decomposed into two simpler sub-problems, wherein the
linearization of the SCAD norm resulted in an adaptively weighted soft thresholding rule with a sparsity enhancing effect. 

23.  K-edge ratio method for identification of multiple nanoparticulate contrast agents by spectral CT imaging
Ghadiri H, Ay MR, Shiran MB, Soltanian-Zadeh H, Zaidi H.
Journal Paper
BRITISH JOURNAL OF RADIOLOGY

 Abstract

Objective: Recently introduced energy-sensitive X-ray CT makes it feasible to discriminate different nanoparticulate contrast materials. The purpose of this work is to present a K-edge ratio method for differentiating multiple simultaneous contrast agents using spectral CT.
Methods: The ratio of two images relevant to energy bins straddling the K-edge of the materials is calculated using an analytic CT simulator. In the resulting parametric map, the selected contrast agent regions can be identified using a thresholding algorithm. The K-edge ratio algorithm is applied to spectral images of simulated phantoms to identify and differentiate up to four simultaneous and targeted CT contrast agents.
Results: We show that different combinations of simultaneous CT contrast agents can be identified by the proposed K-edge ratio method when energy-sensitive CT is used. In the K-edge parametric maps, the pixel values for biological tissues and contrast agents reach a maximum of 0.95, whereas for the selected contrast agents, the pixel values are larger than 1.10. The number of contrast agents that can be discriminated is limited owing to photon starvation. For reliable material discrimination, minimum photon counts corresponding to 140 kVp, 100mAs and 5-mm slice thickness must be used. 
Conclusion: The proposed K-edge ratio method is a straightforward and fast method for identification and discrimination of multiple simultaneous CT contrast agents. 

24.  X-ray CT metal artifact reduction using wavelet domain L0 sparse regularization
Mehranian A , Ay MR, Rahmim A, Zaidi H.
Journal Paper
IEEE TRANSACTIONS ON MEDICAL IMAGING

 Abstract

X-ray computed tomography (CT) imaging of patients with metallic implants usually suffers from streaking metal artifacts. In this paper, we propose a new projection completion metal artifact reduction (MAR) algorithm by formulating the completion of missing projections as a regularized inverse problem in the wavelet domain. The Douglas-Rachford splitting (DRS) algorithm was used to iteratively solve the problem. Two types of prior information were exploited in the algorithm: 1) the sparsity of the wavelet coefficients of CT sinograms in a dictionary of translationinvariant wavelets and 2) the detail wavelet coefficients of a prior sinogram obtained from the forward projection of a segmented CT image. A pseudo- L0 synthesis prior was utilized to exploit and promote the sparsity of wavelet coefficients. The proposed L0-DRS MAR algorithm was compared with standard linear interpolation and the normalized metal artifact reduction (NMAR) approach proposed by Meyer using both simulated and clinical studies including hip prostheses, dental fillings, spine fixation and electroencephalogram electrodes in brain imaging. The qualitative and quantitative evaluations showed that our algorithm substantially suppresses streaking artifacts and can outperform both linear interpolation and NMAR

25.  A novel non-linear recursive filter design for extracting high rate pulse features in nuclear medicine imaging and spectroscopy
Salar Sajedia, Alireza Kamal Asle, Mohammad R. Aya, Mohammad H. Farahania, Arman Rahmimf
Journal Paper
Medical Engineering & Physics

 Abstract

Applications in imaging and spectroscopy rely on pulse processing methods for appropriate data generation. Often, the particular method utilized does not highly impact data quality, whereas in some scenarios, such as in the presence of high count rates or high frequency pulses, this issue merits extra consideration. In the present study, a new approach for pulse processing in nuclear medicine imaging and spectroscopy is introduced and evaluated. The new nonlinear recursive filter (NLRF) performs nonlinear processing of the input signal and extracts the main pulse characteristics, having the powerful ability to recover pulses that would ordinarily result in pulse pile-up. The filter design defines sampling frequencies lower than the Nyquist frequency. In the literature, for systems involving NaI(Tl) detectors and photomultiplier tubes (PMTs), with a signal bandwidth considered as 15 MHz, the sampling frequency should be at least 30 MHz (the Nyquist rate), whereas in the present work, a sampling rate of 3.3 MHz was shown to yield very promising results. This was obtained by exploiting the known shape feature instead of utilizing a general sampling algorithm. The simulation and experimental results show that the proposed filter enhances count rates in spectroscopy. With this filter, the system behaves almost identically as a general pulse detection system with a dead time considerably reduced to the new sampling time (300 ns). Furthermore, because of its unique feature for determining exact event times, the method could prove very useful in time-of-flight PET imaging.

26.  Biodistribution of ultra small superparamagnetic iron oxide nanoparticles in BALB mice
Saeed Shanehsazzadeh • Mohammad Ali Oghabian • Fariba Johari Daha • Massoud Amanlou • Barry J. Allen
Journal Paper
JOURNAL OF RADIOANALYTICAL AND NUCLEAR CHEMISTRY

 Abstract

Recently ultrasmall superparamagnetic iron oxide (USPIO) nanoparticles (NPs) have been widely used for medical applications. One of their important applications is using these particles as MRI contrast agent. While various research works have been done about MRI application of USPIOs, there is limited research about their uptakes in various organs. The aim of this study was to evaluate the biodistribution of dextran coated iron oxide NPs labelled with Tc-99m in various organs via intravenous injection in Balb/c mice. The magnetite NPs were dispersed in phosphate buffered saline and SnCl2 which was used as a reduction reagent. Subsequently, the radioisotope Tc-99m was mixed directly into the reaction solution. The labeling efficiency of USPIOs labeled with Tc-99m, was above 99 %. Sixty mice were sacrificed at 12 different time points (From 1 min to 48 h post injections; five mice at each time).
The percentage of injected dose per gram of each organ was measured by direct counting for 19 harvested organs of the mice. The biodistribution of Tc-99m-USPIO in Balb/c mice showed dramatic uptake in reticuloendothelial system. Accordingly, about 75 percent of injected dose was found in spleen and liver at 15 min post injection. More than 24 % of the NPs remain in liver after 48 h postinjection and their clearance is so fast in other organs. The results suggest that USPIOs as characterized in our study can be potentially used as contrast agent in MR Imaging, distributing reticuloendothelial system specially spleen and liver. 

27.  MRI-guided attenuation correction in whole-body PET/MR:assessment of the effect of bone attenuation
A. Akbarzadeh • M. R. Ay • A. Ahmadian • N. Riahi Alam • H. Zaidi
Journal Paper
ANNALS OF NUCLEAR MEDICINE

 Abstract

Hybrid PET/MRI presents many advantages in comparison with its counterpart PET/CT in terms of improved soft- issue contrast, decrease in radiation exposure, and truly simultaneous and multiparametric imaging capabilities. However, the lack of well-established methodology for MR-based attenuation correction is hampering further development and wider acceptance of this technology. We assess the impact of ignoring bone attenuation and using different tissue classes for generation of the attenuation map on the accuracy of attenuation correction of PET data.
This work was performed using simulation studies based on the XCAT phantom and clinical input data. For the latter, PET and CT images of patients were used as input for the analytic simulation model using realistic activity distributions where CT-based attenuation correction was utilized as reference for comparison. For both phantom and clinical studies, the reference attenuation map was classified into various numbers of tissue classes to produce three (air, soft tissue and lung), four (air, lungs, soft tissue and cortical bones) and five (air, lungs, soft tissue, cortical bones and spongeous bones) class attenuation maps.
The phantom studies demonstrated that ignoring bone increases the relative error by up to 6.8 % in the body and up to 31.0 % for bony regions. Likewise, the simulated clinical studies showed that the mean relative error reached 15 % for lesions located in the body and 30.7 % for lesions located in bones, when neglecting bones. These results demonstrate an underestimation of about 30 % of tracer uptake when neglecting bone, which in turn imposes substantial loss of quantitative accuracy for PET images produced by hybrid PET/MRI systems.
 

Latest News

Articles