Masoom A Haider

Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada

Are you Masoom A Haider?

Claim your profile

Publications (256)790.08 Total impact

  • M. A. Haider · A. Vosough · F. Khalvati · A. Kiss · B. Ganeshan · G. Bjarnason ·

    Radiological Society of North America (RSNA) Annual Meeting; 12/2015
  • [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.
  • F. Khalvati · A. Wong · M. A. Haider ·

    Canadian Cancer Research Conference; 11/2015
  • [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.
    Journal of the American Society of Nephrology 10/2015; DOI:10.1681/ASN.2015060648 · 9.34 Impact Factor
  • [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.
    European Urology 10/2015; DOI:10.1016/j.eururo.2015.08.052 · 13.94 Impact Factor
  • Andrew Cameron · Farzad Khalvati · M. A. Haider · Alexander Wong ·
    [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.
    IEEE Transactions on Biomedical Engineering 09/2015; DOI:10.1109/TBME.2015.2485779 · 2.35 Impact Factor
  • Farzad Khalvati · Shahryar Rahnamayan · Masoom A. Haider · H. R. Tizhoosh ·
    [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.
    Journal of Digital Imaging 09/2015; DOI:10.1007/s10278-015-9844-y · 1.19 Impact Factor
  • Source

    European Urology 09/2015; DOI:10.1016/j.eururo.2015.08.038 · 13.94 Impact Factor
  • [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.
    International journal of radiation oncology, biology, physics 09/2015; DOI:10.1016/j.ijrobp.2015.09.009 · 4.26 Impact Factor
  • [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.
  • [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.
  • 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.
    Magnetic Resonance Imaging 08/2015; DOI:10.1016/j.mri.2015.08.009 · 2.09 Impact Factor
  • [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.
    American Journal of Roentgenology 08/2015; 205(2):W177-84. DOI:10.2214/AJR.14.13098 · 2.73 Impact Factor
  • 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.
    BMC Medical Imaging 07/2015; 15(1). DOI:10.1186/s12880-015-0081-0 · 1.31 Impact Factor
  • A. Chung · C. Scharfenberger · F. Khalvati · A. Wong · M. A. Haider ·

    International Conference on Image Analysis and Recognition (ICIAR), Niagara Falls, ON, Canada; 07/2015
  • Sangeet Ghai · Masoom A Haider ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Multiparametric-magnetic resonance imaging (mp-MRI) has shown promising results in diagnosis, localization, risk stratification and staging of clinically significant prostate cancer. It has also opened up opportunities for focal treatment of prostate cancer. Combinations of T2-weighted imaging, diffusion imaging, perfusion (dynamic contrast-enhanced imaging) and spectroscopic imaging have been used in mp-MRI assessment of prostate cancer, but T2 morphologic assessment and functional assessment by diffusion imaging remains the mainstay for prostate cancer diagnosis on mp-MRI. Because assessment on mp-MRI can be subjective, use of the newly developed standardized reporting Prostate Imaging and Reporting Archiving Data System scoring system and education of specialist radiologists are essential for accurate interpretation. This review focuses on the present status of mp-MRI in prostate cancer and its evolving role in the management of prostate cancer.
    Indian Journal of Urology 07/2015; 31(3):194-201. DOI:10.4103/0970-1591.159606
  • S. M. Kim · M. Milosevic · M. A. Haider · D. Jaffray · I. Yeung ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: In dynamic contrast enhanced CT (DCE-CT) study, a CT scanning with high temporal resolution is necessary to obtain accurate kinetic parameter values, but such scanning scheme substantially increases the radiation dose to the patient. A method of principal component analysis (PCA) filtering combined with the arterial input function (AIF) estimation technique is proposed to reduce patient radiation dose, while maintaining high accuracy of kinetic parameter estimates in a pixel-by-pixel analysis of DCE-CT data acquired at a low scanning frequency. Methods: With the coarsely sampled AIF of a patient, an AIF in high temporal resolution can be generated by using the previously published technique which uses the orthonormal bases of the arterial impulse responses (AIR) extracted from a cohort of 34 patients with cervical cancer. In addition, principal component analysis (PCA) filtering was applied to the tissue curves of all the pixels in a region of interest to increase their signal to noise ratios (SNR). The proposed method was applied to each DCE-CT data set of a cohort of 14 patients at varying levels of down sampling schemes between intervals from 2 to 15 s. Subsequently, the kinetic analyses using the modified Tofts’ model and singular value decomposition (SVD) method were performed for each of the down-sampling schemes. 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 results of kinetic analyses using the proposed method compared with down sampling alone showed that the method is superior in maintaining the accuracy in the quantitative histogram parameters of volume transfer constant (standard deviation (SD), 98th percentile, and range), blood volume fraction (mean, SD, 98th percentile, and range), and blood flow (mean and median) for longer sampling intervals between 10 and 15 s. The preliminary results suggest that the method is able to support the longer scanning intervals at a cost of 8.4%–20.5% loss in accuracy of the histogram parameters for volume transfer constant and 6.9%–18.2% for blood volume fraction, and 5.9%–8.7% for blood flow. Conclusions: The radiation dose to patient during a DCE-CT study can be reduced by up to a factor of 15 with the proposed method of PCA filtering combined with the AIF estimation technique. The results indicate that the method is useful for pixel-by-pixel kinetic analysis of DCE-CT data for patients with cervical cancer.
    2015 World Congress on Medical Physics and Biomedical Engineering, Toronto; 06/2015
  • E. Li · M. J. Sha fiee · Audrey Chung · Farzad Khalvati · Alexander Wong · Masoom A. Haider ·

    International Society for Magnetic Resonance in Medicine (ISMRM), Toronto, ON, Canada; 06/2015
  • C. Scharfenberger · D. Lui · F. Khalvati · A. Wong · M. A. Haider ·

    Annual Meeting of International Society for Magnetic Resonance in Medicine (ISMRM), Toronto, ON, Canada; 06/2015

  • Annual Meeting of Inter- national Society for Magnetic Resonance in Medicine (ISMRM), Toronto, ON, Canada; 06/2015

Publication Stats

4k Citations
790.08 Total Impact Points


  • 2012-2015
    • Sunnybrook Health Sciences Centre
      • Department of Medical Imaging
      Toronto, Ontario, Canada
  • 2000-2015
    • University of Toronto
      • • Department of Medical Imaging
      • • Department of Surgery
      • • Department of Radiation Oncology
      Toronto, Ontario, Canada
  • 2005-2014
    • University Health Network
      • • Joint Department of Medical Imaging
      • • Department of Radiology
      Toronto, Ontario, Canada
  • 2003-2013
    • The Princess Margaret Hospital
      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
  • 2009
    • Illinois Institute of Technology
      • Department of Electrical & Computer Engineering
      Chicago, IL, United States