Masoom A Haider

University of Toronto, Toronto, Ontario, Canada

Are you Masoom A Haider?

Claim your profile

Publications (265)794.5 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: The prognosis for locally advanced esophageal cancer is poor despite the use of trimodality therapy. In this phase II study, we report the feasibility, tolerability and efficacy of adjuvant sunitinib. Included were patients with stage IIa, IIB or III cancer of the thoracic esophagus or gastroesophageal junction. Neoadjuvant therapy involved Irinotecan (65 mg/m(2) ) + Cisplatin (30 mg/m(2) ) on weeks 1 and 2, 4 and 5, 7 and 8 with concurrent radiation (50Gy/25 fractions) on weeks 4-8. Sunitinib was commenced 4-13 weeks after surgery and continued for one year. Sixty-one patients were included in the final analysis, 36 patients commenced adjuvant sunitinib. Fourteen patients discontinued sunitinib due to disease recurrence (39%) within the 12-month period, 12 (33%) discontinued due to toxicity, and 3 (8%) requested cessation of therapy. In the overall population, median survival was 26 months with a 2 and 3-year survival rate of 52% and 35%, respectively. The median survival for the 36 patients treated with sunitinib was 35 months and 2-year survival probability of 68%. In a historical control, a prior phase II study with the same trimodality therapy (n = 43), median survival was 36 months, with a 2-year survival of 67%. Initiation of adjuvant sunitinib is feasible, but poorly tolerated, with no signal of additional benefit over trimodality therapy for locally advanced esophageal cancer.
    No preview · Article · Jan 2016 · Diseases of the Esophagus
  • Sun Mo Kim · Masoom A. Haider · David A. Jaffray · Ivan W. T. Yeung
    [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: A previously proposed method to reduce radiation dose to patient in dynamic contrast-enhanced (DCE) CT is enhanced by principal component analysis (PCA) filtering which improves the signal-to-noise ratio (SNR) of time-concentration curves in the DCE-CT study. The efficacy of the combined method to maintain the accuracy of kinetic parameter estimates at low temporal resolution is investigated with pixel-by-pixel kinetic analysis of DCE-CT data. Methods: The method is based on DCE-CT scanning performed with low temporal resolution to reduce the radiation dose to the patient. The arterial input function (AIF) with high temporal resolution can be generated with a coarsely sampled AIF through a previously published method of AIF estimation. To increase the SNR of time-concentration curves (tissue curves), first, a region-of-interest is segmented into squares composed of 3 × 3 pixels in size. Subsequently, the PCA filtering combined with a fraction of residual information criterion is applied to all the segmented squares for further improvement of their SNRs. The proposed method was applied to each DCE-CT data set of a cohort of 14 patients at varying levels of down-sampling. The kinetic analyses using the modified Tofts' model and singular value decomposition method, then, were carried out for each of the down-sampling schemes between the intervals from 2 to 15 s. The results were compared with analyses done with the measured data in high temporal resolution (i.e., original scanning frequency) as the reference. Results: The patients' AIFs were estimated to high accuracy based on the 11 orthonormal bases of arterial impulse responses established in the previous paper. In addition, noise in the images was effectively reduced by using five principal components of the tissue curves for filtering. Kinetic analyses using the proposed method showed superior results compared to those with down-sampling alone; they were able to maintain the accuracy in the quantitative histogram parameters of volume transfer constant [standard deviation (SD), 98th percentile, and range], rate constant (SD), blood volume fraction (mean, SD, 98th percentile, and range), and blood flow (mean, SD, median, 98th percentile, and range) for sampling intervals between 10 and 15 s. Conclusions: The proposed method of PCA filtering combined with the AIF estimation technique allows low frequency scanning for DCE-CT study to reduce patient radiation dose. The results indicate that the method is useful in pixel-by-pixel kinetic analysis of DCE-CT data for patients with cervical cancer.
    No preview · Article · Jan 2016 · Medical Physics
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Magnetic Resonance Imaging (MRI) is a crucial medical imaging technology for the screening and diagnosis of frequently occurring cancers. However image quality may suffer by long acquisition times for MRIs due to patient motion, as well as result in great patient discomfort. Reducing MRI acquisition time can reduce patient discomfort and as a result reduces motion artifacts from the acquisition process. Compressive sensing strategies, when applied to MRI, have been demonstrated to be effective at decreasing acquisition times significantly by sparsely sampling the \emph{k}-space during the acquisition process. However, such a strategy requires advanced reconstruction algorithms to produce high quality and reliable images from compressive sensing MRI. This paper proposes a new reconstruction approach based on cross-domain stochastically fully connected conditional random fields (CD-SFCRF) for compressive sensing MRI. The CD-SFCRF introduces constraints in both \emph{k}-space and spatial domains within a stochastically fully connected graphical model to produce improved MRI reconstruction. Experimental results using T2-weighted (T2w) imaging and diffusion-weighted imaging (DWI) of the prostate show strong performance in preserving fine details and tissue structures in the reconstructed images when compared to other tested methods even at low sampling rates.
    Full-text · Article · Dec 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Diffusion weighted magnetic resonance imaging (DW-MRI) is a powerful tool in imaging-based prostate cancer (PCa) screening and detection. Endorectal coils are commonly used in DW-MRI to improve the signal-to-noise ratio (SNR) of the acquisition, at the expense of significant intensity inhomogeneities (bias field) that worsens as we move away from the endorectal coil. The presence of bias field can have a significant negative impact on the accuracy of different image analysis tasks, as well as the accuracy of PCa tumor localization, thus leading to increased inter- and intra-observer variability. The previously proposed bias field correction methods often suffer from undesired noise amplification that can reduce the image quality of the resulting bias-corrected DW-MRI data. Here, we propose a unified data reconstruction approach that enables joint compensation of bias field as well as data noise in diffusion weighted endorectal magnetic resonance (DW-EMR) imaging. The proposed noise-compensated, bias-corrected (NCBC) data reconstruction method takes advantage of a novel stochastically fully connected joint conditional random field (SFC-JCRF) model to mitigate the effects of data noise and bias field in the reconstructed DW-EMR prostate imaging data. The proposed NCBC reconstruction method was tested on synthetic DW-EMR data, physical DW-EMR phantom, as well as real DW-EMR imaging data. Both qualitative and quantitative analysis illustrated that the proposed NCBC method can achieve improved image quality when compared to other tested bias correction methods. As such, the proposed NCBC method can have strong potential for improving the consistency of image interpretations, thus leading to more accurate PCa diagnosis.
    Full-text · Article · Dec 2015
  • M. A. Haider · A. Vosough · F. Khalvati · A. Kiss · B. Ganeshan · G. Bjarnason

    No preview · Conference Paper · Dec 2015
  • [Show abstract] [Hide abstract]
    ABSTRACT: Radiomics has proven to be a powerful prognostic tool for cancer detection, and has previously been applied in lung, breast, prostate, and head-and-neck cancer studies with great success. However, these radiomics-driven methods rely on pre-defined, hand-crafted radiomic feature sets that can limit their ability to characterize unique cancer traits. In this study, we introduce a novel discovery radiomics framework where we directly discover custom radiomic features from the wealth of available medical imaging data. In particular, we leverage novel StochasticNet radiomic sequencers for extracting custom radiomic features tailored for characterizing unique cancer tissue phenotype. Using StochasticNet radiomic sequencers discovered using a wealth of lung CT data, we perform binary classification on 42,340 lung lesions obtained from the CT scans of 93 patients in the LIDC-IDRI dataset. Preliminary results show significant improvement over previous state-of-the-art methods, indicating the potential of the proposed discovery radiomics framework for improving cancer screening and diagnosis.
    No preview · Article · Nov 2015
  • F. Khalvati · A. Wong · M. A. Haider

    No preview · Conference Paper · Nov 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we describe the underlying methodology behind discoveryradiomics, where the ultimate goal is to discover customized,abstract radiomic feature models directly from the wealth of medicalimaging data to better capture highly unique tumor traits beyondwhat can be captured using hand-crafted radiomic featuremodels. We further explore the current state-of-the-art in discoveryradiomics and their application to various forms of cancer suchas prostate cancer and lung cancer, and show that discovery radiomicscan yield significant potential clinical impact.
    Full-text · Article · Oct 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a novel compensated diffusion magnetic resonanceimaging (cdMRI) system for improved tissue detail and contrastfor screening and diagnosis. The proposed cdMRI systemincorporates the intrinsic properties of the MRI apparatus to compensatefor artifacts and degradation caused through the imagingprocess. Experimental results for prostate imaging show that significantimprovements in tissue detail and contrast can be obtainedcompared to current MRI systems.
    Full-text · Article · Oct 2015
  • [Show abstract] [Hide abstract]
    ABSTRACT: Renal disease variability in autosomal dominant polycystic kidney disease (ADPKD) is strongly influenced by the gene locus (PKD1 versus PKD2). Recent studies identified nontruncating PKD1 mutations in approximately 30% of patients who underwent comprehensive mutation screening, but the clinical significance of these mutations is not well defined. We examined the genotype-renal function correlation in a prospective cohort of 220 unrelated ADPKD families ascertained through probands with serum creatinine ≤1.4 mg/dl at recruitment. We screened these families for PKD1 and PKD2 mutations and reviewed the clinical outcomes of the probands and affected family members. Height-adjusted total kidney volume (htTKV) was obtained in 161 affected subjects. Multivariate Cox proportional hazard modeling for renal and patient survival was performed in 707 affected probands and family members. Overall, we identified pathogenic mutations in 84.5% of our families, in which the prevalence of PKD1 truncating, PKD1 in-frame insertion/deletion, PKD1 nontruncating, and PKD2 mutations was 38.3%, 4.3%, 27.1%, and 30.3%, respectively. Compared with patients with PKD1 truncating mutations, patients with PKD1 in-frame insertion/deletion, PKD1 nontruncating, or PKD2 mutations have smaller htTKV and reduced risks (hazard ratio [95% confidence interval]) of ESRD (0.35 [0.14 to 0.91], 0.10 [0.05 to 0.18], and 0.03 [0.01 to 0.05], respectively) and death (0.31 [0.11 to 0.87], 0.20 [0.11 to 0.38], and 0.18 [0.11 to 0.31], respectively). Refined genotype-renal disease correlation coupled with targeted next generation sequencing of PKD1 and PKD2 may provide useful clinical prognostication for ADPKD.
    No preview · Article · Oct 2015 · Journal of the American Society of Nephrology
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The Prostate Imaging - Reporting and Data System Version 2 (PI-RADS™ v2) is the product of an international collaboration of the American College of Radiology (ACR), European Society of Uroradiology (ESUR), and AdMetech Foundation. It is designed to promote global standardization and diminish variation in the acquisition, interpretation, and reporting of prostate multiparametric magnetic resonance imaging (mpMRI) examination, and it is based on the best available evidence and expert consensus opinion. It establishes minimum acceptable technical parameters for prostate mpMRI, simplifies and standardizes terminology and content of reports, and provides assessment categories that summarize levels of suspicion or risk of clinically significant prostate cancer that can be used to assist selection of patients for biopsies and management. It is intended to be used in routine clinical practice and also to facilitate data collection and outcome monitoring for research.
    Full-text · Article · Oct 2015 · European Urology
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a quantitative radiomics feature model for performing prostate cancer detection using Multi- Parametric MRI (mpMRI). It incorporates a novel tumour candidate identification algorithm to efficiently and thoroughly identify regions of concern and constructs a comprehensive radiomics feature model to detect tumourous regions. In contrast to conventional automated classification schemes, this radiomicsbased feature model aims to ground its decisions in a way that can be interpreted and understood by the diagnostician. This is done by grouping features into high-level feature categories which are already used by radiologists to diagnose prostate cancer: Morphology, Asymmetry, Physiology, and Size (MAPS), using biomarkers inspired by the PI-RADS guidelines for performing structured reporting on prostate MRI. Clinical mpMRI data were collected from thirteen men with histology-confirmed prostate cancer and labeled by an experienced radiologist. These annotated data were used to train classifiers using the proposed radiomics-driven feature model in order to evaluate the classification performance. The preliminary experimental results indicated that the proposed model outperformed each of its constituent feature groups as well as a comparable conventional mpMRI feature model. A further validation of the proposed algorithm will be conducted using a larger dataset as future work.
    No preview · Article · Sep 2015 · IEEE Transactions on Biomedical Engineering
  • [Show abstract] [Hide abstract]
    ABSTRACT: Accurate and fast segmentation and volume estimation of the prostate gland in magnetic resonance (MR) images are necessary steps in the diagnosis, treatment, and monitoring of prostate cancer. This paper presents an algorithm for the prostate gland volume estimation based on the semi-automated segmentation of individual slices in T2-weighted MR image sequences. The proposed sequential registration-based segmentation (SRS) algorithm, which was inspired by the clinical workflow during medical image contouring, relies on inter-slice image registration and user interaction/correction to segment the prostate gland without the use of an anatomical atlas. It automatically generates contours for each slice using a registration algorithm, provided that the user edits and approves the marking in some previous slices. We conducted comprehensive experiments to measure the performance of the proposed algorithm using three registration methods (i.e., rigid, affine, and nonrigid). Five radiation oncologists participated in the study where they contoured the prostate MR (T2-weighted) images of 15 patients both manually and using the SRS algorithm. Compared to the manual segmentation, on average, the SRS algorithm reduced the contouring time by 62 % (a speedup factor of 2.64×) while maintaining the segmentation accuracy at the same level as the intra-user agreement level (i.e., Dice similarity coefficient of 91 versus 90 %). The proposed algorithm exploits the inter-slice similarity of volumetric MR image series to achieve highly accurate results while significantly reducing the contouring time.
    No preview · Article · Sep 2015 · Journal of Digital Imaging
  • Source

    Full-text · Article · Sep 2015 · European Urology
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: Preclinical studies have shown that angiogenesis inhibition can improve response to radiation therapy (RT). The purpose of this phase 1 study was to examine the angiogenesis inhibitor sorafenib in patients with cervical cancer receiving radical RT and concurrent cisplatin (RTCT). Methods and materials: Thirteen patients with stage IB to IIIB cervical cancer participated. Sorafenib was administered daily for 7 days before the start of standard RTCT in patients with early-stage, low-risk disease and also during RTCT in patients with high-risk disease. Biomarkers of tumor vascularity, perfusion, and hypoxia were measured at baseline and again after 7 days of sorafenib alone before the start of RTCT. The median follow-up time was 4.5 years. Results: Initial complete response was seen in 12 patients. One patient died without achieving disease control, and 4 experienced recurrent disease. One patient with an extensive, infiltrative tumor experienced pelvic fistulas during treatment. The 4-year actuarial survival was 85%. Late grade 3 gastrointestinal toxicity developed in 4 patients. Sorafenib alone produced a reduction in tumor perfusion/permeability and an increase in hypoxia, which resulted in early closure of the study. Conclusions: Sorafenib increased tumor hypoxia, raising concern that it might impair rather than improve disease control when added to RTCT.
    No preview · Article · Sep 2015 · International journal of radiation oncology, biology, physics
  • [Show abstract] [Hide abstract]
    ABSTRACT: Prostate cancer is the most diagnosed form of cancer in Canadian men, and is the third leading cause of cancer death. Despite these statistics, prognosis is relatively good with a sufficiently early diagnosis, making fast and reliable prostate cancer detection crucial. As imaging-based prostate cancer screening, such as magnetic resonance imaging (MRI), requires an experienced medical professional to extensively review the data and perform a diagnosis, radiomics-driven methods help streamline the process and has the potential to significantly improve diagnostic accuracy and efficiency, and thus improving patient survival rates. These radiomics-driven methods currently rely on hand-crafted sets of quantitative imaging-based features, which are selected manually and can limit their ability to fully characterize unique prostate cancer tumour phenotype. In this study, we propose a novel \textit{discovery radiomics} framework for generating custom radiomic sequences tailored for prostate cancer detection. Discovery radiomics aims to uncover abstract imaging-based features that capture highly unique tumour traits and characteristics beyond what can be captured using predefined feature models. In this paper, we discover new custom radiomic sequencers for generating new prostate radiomic sequences using multi-parametric MRI data. We evaluated the performance of the discovered radiomic sequencer against a state-of-the-art hand-crafted radiomic sequencer for computer-aided prostate cancer detection with a feedforward neural network using real clinical prostate multi-parametric MRI data. Results for the discovered radiomic sequencer demonstrate good performance in prostate cancer detection and clinical decision support relative to the hand-crafted radiomic sequencer. The use of discovery radiomics shows potential for more efficient and reliable automatic prostate cancer detection.
    No preview · Article · Aug 2015
  • [Show abstract] [Hide abstract]
    ABSTRACT: Lung cancer is one of the most diagnosed form of cancer in the world and the leading cause for cancer related deaths. A powerful tool that can aid radiologists in delivering more accurate and faster diagnosis is radiomics, where a wealth of quantitative imaging features are derived from imaging data for characterizing tumour phenotype and for quantitative diagnosis. In this study, we take the idea of radiomics one step further by introducing the concept of discovery radiomics for lung cancer detection using CT imaging data. Rather than using pre-defined, hand-engineered feature models as with current radiomics-driven methods, we discover custom radiomic sequencers that can generate radiomic sequences consisting of abstract imaging-based features tailored for characterizing lung tumour phenotype. In this study, we realize these custom radiomic sequencers as deep convolutional sequencers, and discover such sequencers using a deep convolutional neural network learning architecture directly based on a wealth of CT imaging data. To illustrate the prognostic power and effectiveness of the radiomic sequences produced by the discovered sequencer, we perform a classification between malignant and benign lung lesions using helical lung CT scans captured from 93 patients with diagnostic data from the LIDC-IDRI dataset. Using the clinically provided diagnostic data as ground truth, classification using the discovered radiomic sequencer provided an average accuracy of 77.52% via 10-fold cross-validation with a sensitivity of 79.06% and specificity of 76.11%. These results illustrate the potential for the proposed discovery radiomics approach in aiding radiologists in improving screening efficiency and accuracy.
    No preview · Article · Aug 2015
  • Hatef Mehrabian · Michael Da Rosa · Masoom A Haider · Anne L Martel
    [Show abstract] [Hide abstract]
    ABSTRACT: Dynamic contrast enhanced (DCE)-MRI combined with pharmacokinetic (PK) modeling of a tumor provides information about its perfusion and vascular permeability. Most PK models require the time course of contrast agent concentration in blood plasma as an input, which cannot be measured directly at the tissue of interest, and is approximated with an arterial input function (AIF). Variability in methods used in estimating the AIF and inter-observer variability in region of interest selection are major sources of discrepancy between different studies. This study had two aims. The first was to determine whether a local vascular input function estimated using an adaptive complex independent component analysis (AC-ICA) algorithm could be used to estimate PK parameters from clinical dynamic contrast enhanced (DCE)-MRI studies. The second aim was to determine whether normalizing the input function using the area under the curve would improve the results of PK analysis. AC-ICA was applied to DCE-MRI of 27 prostate cancer patients and the intravascular signal was estimated. This signal was converted into contrast agent concentration to give a local vascular input function (VIF) which was used as the input function for PK analysis. We compared K(trans) values for normal peripheral zone (PZ) and tumor tissues using the local VIF with those obtained using a conventional AIF obtained from the femoral artery. We also compared the K(trans) values obtained from the un-normalized input functions with the KN(trans) values obtained after normalizing the AIF and local VIF. Normalization of the input function resulted in smaller variation in PK parameters (KN(trans) vs. K(trans)for normal PZ tissue was 0.20±0.04 mM.min(-1) vs. 0.87±0.54 min(-1) for local VIF and 0.21±0.07 mM.min(-1) vs. 0.25±0.29 min(-1) for AIF) and better separation of the normal and tumor tissues (effect-size of this separation using KN(trans) vs. K(trans) was 0.89 vs. 0.75 for local VIF and 0.94 vs. 0.41 for AIF). The AC-ICA and AIF-based analyses provided similar (KN(trans)) values in normal PZ tissue of prostate across patients. Normalizing the input function before PK analysis significantly improved the reproducibility of the PK parameters and increased the separation between normal and tumor tissues. Using AC-ICA allows a local VIF to be estimated and the resulting PK parameters are similar to those obtained using a more conventional AIF; this may be valuable in studies where an artery is not available in the field of view. Copyright © 2015. Published by Elsevier Inc.
    No preview · Article · Aug 2015 · Magnetic Resonance Imaging
  • [Show abstract] [Hide abstract]
    ABSTRACT: Focal therapy is an emerging approach to the treatment of localized prostate cancer. The purpose of this study was to report the 6-month follow-up oncologic and functional data of the initial phase 1 trial of patients treated with focal transrectal MRI-guided focused ultrasound in North America. Four patients with a prostate-specific antigen (PSA) level of 10 ng/mL or less, tumor classification cT2a or less, and a Gleason score of 6 (3 + 3) were prospectively enrolled in the study and underwent multiparametric MRI and transrectal ultrasound-guided prostate systematic biopsy. Under MRI guidance and real-time monitoring with MR thermography, focused high-frequency ultrasound energy was delivered to ablate the target tissue. The incidence and severity of treatment-related adverse events were recorded along with responses to serial quality-of-life questionnaires for 6 months after treatment. Oncologic outcomes were evaluated with multiparametric MRI and repeat transrectal ultrasound-guided biopsy 6 months after treatment. Four patients with a total of six target lesions were treated and had complications graded Clavien-Dindo I or less. Quality-of-life parameters were similar between baseline and 6-months. All four patients had normal MRI findings in the treated regions (100%), biopsy showed that three patients (75%) were clear of disease in the treated regions, representing complete ablation of five target lesions (83%). All patients had at least one Gleason 6-positive core outside of the treated zone. MRI-guided focused ultrasound is a feasible method of noninvasively ablating low-risk prostate cancers with low morbidity. Further investigation and follow-up are warranted in a larger patient series with appropriate statistical analysis of oncologic and functional outcome measures.
    No preview · Article · Aug 2015 · American Journal of Roentgenology
  • Source
    Dorothy Lui · Amen Modhafar · Masoom A. Haider · Alexander Wong
    [Show abstract] [Hide abstract]
    ABSTRACT: Prostate cancer is one of the most common forms of cancer found in males making early diagnosis important. Magnetic resonance imaging (MRI) has been useful in visualizing and localizing tumor candidates and with the use of endorectal coils (ERC), the signal-to-noise ratio (SNR) can be improved. The coils introduce intensity inhomogeneities and the surface coil intensity correction built into MRI scanners is used to reduce these inhomogeneities. However, the correction typically performed at the MRI scanner level leads to noise amplification and noise level variations. In this study, we introduce a new Monte Carlo-based noise compensation approach for coil intensity corrected endorectal MRI which allows for effective noise compensation and preservation of details within the prostate. The approach accounts for the ERC SNR profile via a spatially-adaptive noise model for correcting non-stationary noise variations. Such a method is useful particularly for improving the image quality of coil intensity corrected endorectal MRI data performed at the MRI scanner level and when the original raw data is not available. SNR and contrast-to-noise ratio (CNR) analysis in patient experiments demonstrate an average improvement of 11.7 and 11.2 dB respectively over uncorrected endorectal MRI, and provides strong performance when compared to existing approaches. Experimental results using both phantom and patient data showed that ACER provided strong performance in terms of SNR, CNR, edge preservation, subjective scoring when compared to a number of existing approaches. A new noise compensation method was developed for the purpose of improving the quality of coil intensity corrected endorectal MRI data performed at the MRI scanner level. We illustrate that promising noise compensation performance can be achieved for the proposed approach, which is particularly important for processing coil intensity corrected endorectal MRI data performed at the MRI scanner level and when the original raw data is not available.
    Full-text · Article · Jul 2015 · BMC Medical Imaging

Publication Stats

5k Citations
794.50 Total Impact Points

Institutions

  • 2000-2016
    • University of Toronto
      • • Department of Medical Imaging
      • • Department of Surgery
      • • Department of Radiation Oncology
      Toronto, Ontario, Canada
  • 2012-2015
    • Sunnybrook Health Sciences Centre
      • Department of Medical Imaging
      Toronto, Ontario, Canada
  • 2003-2013
    • The Princess Margaret Hospital
      Toronto, Ontario, Canada
  • 2005-2012
    • University Health Network
      • • Department of Radiology
      • • Joint Department of Medical Imaging
      Toronto, Ontario, Canada
  • 2002-2012
    • Mount Sinai Hospital, Toronto
      • Department of Medical Imaging
      Toronto, Ontario, Canada
  • 2010
    • Mount Sinai Hospital
      New York City, New York, United States