A5100675654- Glioma Diagnosis and Treatment
- Radiomics and Machine Learning in Medical Imaging
- Advanced MRI Techniques and Applications
- Brain Metastases and Treatment
- Vascular Malformations Diagnosis and Treatment
- MRI in cancer diagnosis
- Advanced Neuroimaging Techniques and Applications
- Brain Tumor Detection and Classification
- Intracranial Aneurysms: Treatment and Complications
- Fetal and Pediatric Neurological Disorders
- Immune cells in cancer
- Medical Imaging Techniques and Applications
- Cancer, Hypoxia, and Metabolism
- Meningioma and schwannoma management
- Cerebrospinal fluid and hydrocephalus
- Intracerebral and Subarachnoid Hemorrhage Research
- Functional Brain Connectivity Studies
- Acute Ischemic Stroke Management
- Lanthanide and Transition Metal Complexes
- Vascular Malformations and Hemangiomas
- Cancer Immunotherapy and Biomarkers
- Neurosurgical Procedures and Complications
- Cerebrovascular and Carotid Artery Diseases
- Management of metastatic bone disease
- Nanoplatforms for cancer theranostics
Stanford University
2015-2024
Palo Alto University
2022-2024
Stanford Health Care
2017-2024
Stanford Medicine
2015-2024
Weatherford College
2023
Stanford Cancer Institute
2016-2022
Cancer Prevention Institute of California
2016-2022
Lucile Packard Children's Hospital
2013-2021
Microsoft (Israel)
2017
Santa Clara Valley Medical Center
2011-2015
The BraTS 2021 challenge celebrates its 10th anniversary and is jointly organized by the Radiological Society of North America (RSNA), American Neuroradiology (ASNR), Medical Image Computing Computer Assisted Interventions (MICCAI) society. Since inception, has been focusing on being a common benchmarking venue for brain glioma segmentation algorithms, with well-curated multi-institutional multi-parametric magnetic resonance imaging (mpMRI) data. Gliomas are most primary malignancies central...
Detecting and segmenting brain metastases is a tedious time-consuming task for many radiologists, particularly with the growing use of multi-sequence 3D imaging. This study demonstrates automated detection segmentation on MRI using deep learning approach based fully convolution neural network (CNN). In this retrospective study, total 156 patients from several primary cancers were included. Pre-therapy MR images (1.5T 3T) included pre- post-gadolinium T1-weighted fast spin echo, axial...
Distinct molecular subgroups of pediatric medulloblastoma confer important differences in prognosis and therapy. Currently, tissue sampling is the only method to obtain information for classification. Our goal was develop validate radiomic machine learning approaches predicting medulloblastoma.In this multi-institutional retrospective study, we evaluated MR imaging datasets 109 patients with from 3 children's hospitals January 2001 2014. A computational framework developed extract...
Abstract Purpose: Preclinical studies have demonstrated that postirradiation tumor revascularization is dependent on a stromal cell–derived factor-1 (SDF-1)/C-X-C chemokine receptor type 4 (CXCR4)-driven process in which myeloid cells are recruited from bone marrow. Blocking this axis results survival improvement preclinical models of solid tumors, including glioblastoma (GBM). We conducted phase I/II study to determine the safety and efficacy Macrophage Exclusion after Radiation Therapy...
Readout-segmented echo-planar imaging (EPI) has been suggested as an alternative to single-shot EPI for diffusion-weighted (DWI) with reduced distortion. However, clinical comparisons of readout-segmented and DWI are limited by unmatched parameters reconstruction procedures. Our goal was compare the utility generalized autocalibrating partial parallel acquisition (GRAPPA)-accelerated GRAPPA-accelerated visualization pediatric brain in regions prone distortion, such orbit, skull base,...
Purpose To investigate ferumoxytol-enhanced MRI as a noninvasive imaging biomarker of macrophages in adults with high-grade gliomas. Materials and Methods In this prospective study, gliomas were enrolled between July 2015 2017. Each participant was administered intravenous ferumoxytol (5 mg/kg) underwent 3.0-T 24 hours later. Two sites each tumor selected for intraoperative sampling on the basis degree ferumoxytol-induced signal change. Susceptibility relaxation rates R2* (1/T2*) R2 (1/T2)...
<h3>BACKGROUND AND PURPOSE:</h3> Posterior fossa tumors are the most common pediatric brain tumors. MR imaging is key to tumor detection, diagnosis, and therapy guidance. We sought develop an imaging–based deep learning model for posterior detection pathology classification. <h3>MATERIALS METHODS:</h3> The study cohort comprised 617 children (median age, 92 months; 56% males) from 5 institutions with tumors: diffuse midline glioma of pons (<i>n</i> = 122), medulloblastoma 272), pilocytic...
Abstract The purpose of this study was to assess the clinical value a deep learning (DL) model for automatic detection and segmentation brain metastases, in which neural network is trained on four distinct MRI sequences using an input-level dropout layer, thus simulating scenario missing by training full set all possible subsets input data. This retrospective, multicenter study, evaluated 165 patients with metastases. proposed (ILD) multisequence from 100 validated/tested 10/55 patients,...
Introduction Management of patients with brain metastases is often based on manual lesion detection and segmentation by an expert reader. This a time- labor-intensive process, to that end, this work proposes end-to-end deep learning network for varying number available MRI sequences. Methods We adapt evaluate 2.5D 3D convolution neural trained tested retrospective multinational study from two independent centers, in addition, nnU-Net was adapted as comparative benchmark. Segmentation...
: First-line therapy for high-grade gliomas (HGGs) includes maximal safe surgical resection. The extent of resection predicts overall survival, but current neuroimaging approaches lack tumor specificity. epidermal growth factor receptor (EGFR) is a highly expressed HGG biomarker. We evaluated the safety and feasibility an anti-EGFR antibody, panitumuab-IRDye800, at subtherapeutic doses as imaging agent HGG.
To address the current lack of dynamic susceptibility contrast magnetic resonance imaging (DSC–MRI)-based radiomics to predict isocitrate dehydrogenase (IDH) mutations in gliomas, we present a multicenter study that featured an independent exploratory set for model development and external validation using two cohorts. The maximum performance IDH mutation status prediction on had accuracy 0.544 (Cohen’s kappa: 0.145, F1-score: 0.415, area under curve-AUC: 0.639, sensitivity: 0.733,...
ABSTRACT BACKGROUND AND PURPOSE To investigate the frequency and characteristics of developmental venous anomaly (DVA)‐associated perfusion abnormalities on arterial spin labeling (ASL) bolus perfusion‐weighted imaging (PWI) discuss their potential causes. METHODS We reviewed brain MR reports to identify all DVAs reported studies performed between 2009 2012. DVA location findings PWI and/or ASL were assessed by visual inspection. Sizes categorized as small (<15 mm), medium (15‐25 large...
Fractional tumor burden better correlates with histologic volume fraction in treated glioblastoma than other perfusion metrics such as relative CBV. We defined fractional classes low and high blood to distinguish from treatment effect determine whether can inform treatment-related decision-making.Forty-seven patients high-grade gliomas (primarily glioblastoma) recurrent contrast-enhancing lesions on DSC-MR imaging were retrospectively evaluated after surgical sampling. Histopathologic...
Magnetic resonance images (MRI) of the brain exhibit high dimensionality that pose significant challenges for computational analysis. While models proposed MRIs analyses yield encouraging results, complexity neuroimaging data hinders generalizability and clinical application. We introduce DUNE, a neuroimaging-oriented encoder designed to extract deep-features from multisequence MRIs, thereby enabling their processing by basic machine learning algorithms. A UNet-based autoencoder was trained...