A5006747947- Medical Image Segmentation Techniques
- Radiomics and Machine Learning in Medical Imaging
- AI in cancer detection
- Advanced Neuroimaging Techniques and Applications
- Fetal and Pediatric Neurological Disorders
- Medical Imaging Techniques and Applications
- MRI in cancer diagnosis
- Robotics and Sensor-Based Localization
- Breast Cancer Treatment Studies
- Medical Imaging and Analysis
- Advanced MRI Techniques and Applications
- Parkinson's Disease Mechanisms and Treatments
- Digital Imaging for Blood Diseases
- Advanced Neural Network Applications
- 3D Shape Modeling and Analysis
- Network Traffic and Congestion Control
- Neurological disorders and treatments
- Neonatal and fetal brain pathology
- Intravenous Infusion Technology and Safety
- Cancer Genomics and Diagnostics
- Advanced Queuing Theory Analysis
- Simulation Techniques and Applications
- Voice and Speech Disorders
- Botulinum Toxin and Related Neurological Disorders
- Cell Image Analysis Techniques
Boston Children's Hospital
2016-2021
Icahn School of Medicine at Mount Sinai
2018-2020
Mount Sinai Hospital
2018-2019
Harvard University
2016-2018
Western University
2014-2015
Robarts Clinical Trials
2014-2015
McMaster University
2009-2014
Sharif University of Technology
2007
Breast cancer is the most common invasive in women, affecting more than 10% of women worldwide. Microscopic analysis a biopsy remains one important methods to diagnose type breast cancer. This requires specialized by pathologists, task that i) highly time- and cost-consuming ii) often leads nonconsensual results. The relevance potential automatic classification algorithms using hematoxylin-eosin stained histopathological images has already been demonstrated, but reported results are still...
Pathology Artificial Intelligence Platform (PAIP) is a free research platform in support of pathological artificial intelligence (AI). The main goal the to construct high-quality pathology learning data set that will allow greater accessibility. PAIP Liver Cancer Segmentation Challenge, organized conjunction with Medical Image Computing and Computer Assisted Intervention Society (MICCAI 2019), first image analysis challenge apply datasets. was evaluate new existing algorithms for automated...
This work proposes a novel approach for motion-robust diffusion-weighted (DW) brain MRI reconstruction through tracking temporal head motion using slice-to-volume registration. The slice-level is estimated filtering that allows the during scan and correcting out-of-plane inconsistency in acquired images. Diffusion-sensitized image slices are registered to base volume sequentially over time acquisition order where an outlier-robust Kalman filter, coupled with registration, estimates...
Breast cancer (BC) grading plays a critical role in patient management despite the considerable inter- and intra-observer variability, highlighting need for decision support tools to improve reproducibility prognostic accuracy use clinical practice. The objective was evaluate ability of digital artificial intelligence (AI) assay (PDxBr) enrich BC risk categorization predicting recurrence.In our population-based longitudinal development validation study, we enrolled 2075 patients from Mount...
We present a parallel implementation of new deformable image registration algorithm using the Computer Unified Device Architecture (CUDA). The co-registers preoperative and intraoperative 3-dimensional magnetic resonance (MR) images deforming organ. It employs linear elastic dynamic finite-element model deformation distance measures such as mutual information sum squared differences to align volumetric data sets. Computationally intensive elements method interpolation, displacement force...
Purpose To achieve motion‐robust diffusion compartment imaging (DCI) in near continuously moving subjects based on simultaneous multi‐slice, diffusion‐weighted brain MRI. Methods Simultaneous multi‐slice (SMS) acquisition enables fast and dense sampling of k ‐ q ‐space. We propose to DCI via slice‐level motion correction by exploiting the rigid coupling between simultaneously acquired slices. This provides 3D coverage anatomy that substantially constraints slice‐to‐volume alignment problem....
A unified framework for automatic non-rigid 3D-3D and 3D-2D registration of medical images with static dynamic deformations is proposed in this paper. The problem image approached as a classical state estimation using generic deformation model the soft tissue. technique employs linear elastic continuum mechanics tissue deformation, which discretized finite element method. In method, achieved through Kalman-like filtering process, incorporates information from vector observation prediction...
Real-time registration of pre-operative magnetic resonance (MR) or computed tomography (CT) images with intra-operative Ultrasound (US) can be a valuable tool in image-guided therapies and interventions. This paper presents an automatic method for dynamically tracking the deformation soft tissue based on registering three-dimensional (3D) MR to two-dimensional (2D) US images. The algorithm is concepts state estimation where dynamic finite element (FE)- linear elastic model correlates imaging...
A method is proposed for automatic registration of 3D preoperative magnetic resonance images deformable tissue to a sequence its 2D intraoperative images. The algorithm employs dynamic continuum mechanics model the deformation and similarity (distance) measures such as correlation ratio, mutual information or sum squared differences registration. solely based on present in does not require fiducial markers, feature extraction image segmentation. Results experiments with biopsy training...
Magnetic resonance imaging (MRI) is being increasingly used for image-guided biopsy and focal therapy of prostate cancer. A combined rigid deformable registration technique proposed to register pre-treatment diagnostic 3T magnetic (MR) images, with the identified target tumor(s), intra-treatment 1.5T MR images. The images are acquired patients in strictly supine position using an endorectal coil, while obtained intra-operatively just before insertion ablation needle lithotomy position. An...
Due to the excellent properties of model predictive controllers (MPC) in implementing on nonlinear and time varying systems, utilizing these as Active Queue Management (AQM) strategy is proposed for congestion control computer networks. However, high computational demand solve optimization problem exist a major obstacle when they are applied fast large-scale constrained systems such Small signal linearized TCP/AQM network used design MPC controller then neural trained approximate strategy....
Abstract Population averaged diffusion atlases can be utilized to characterize complex microstructural changes with less bias than data from individual subjects. In this study, a fetal tensor imaging (DTI) atlas was used investigate tract-based in anisotropy and diffusivity vivo 23 38 weeks of gestational age (GA). Healthy pregnant volunteers typically developing fetuses were imaged at 3 Tesla. Acquisition included structural images processed super-resolution algorithm DTI motion-tracked...
A non-rigid registration method is presented for the alignment of pre-procedural magnetic resonance (MR) images with delineated suspicious regions to intra-procedural 3D transrectal ultrasound (TRUS) in TRUS-guided prostate biopsy. In first step, MR and TRUS are aligned rigidly using six pairs manually identified approximate matching points on boundary prostate. Then, two image volumes non-rigidly registered a finite element (FEM)-based linear elastic deformation model. vector observation...
Abstract Background: Genomic testing such as OncotypeDx remains an important component of the treatmentdecision process for many breast cancer (BC) patients. Evidence from Sparano et al. JCO 2021;39:557-564 demonstrated importance combining clinical features tumor grade, size and age with 21-gene recurrence score (i.e., RSClin). Given challenges associated reliability BC grading a prognostic feature, we sought to develop broadly accessible AI-digital test (PDxBr) which included AI-grade...