Zhongnan Fang, PhD
Principal Machine Learning Scientist
AI Evaluation Lab
Radiology, School of Medicine
Stanford University
zhongnanf [at] stanford [.] edu
[Google Scholar]
I am a Principal Machine Learning Scientist at the AI Evaluation Lab of
Stanford University,
Department of Radiology.
Previously, I was a founding member and the Data Science Lead at
LVIS corporation, a Palo Alto based start-up innovating how
neurological diseases are treated with cutting-edge neuroscience findings and AI. I
received my Ph.D. in Electrical Engineering from Stanford University, M.S. from the University of
California Los Angeles, and B.S. from Zhejiang University, China.
My research interests lie in driving innovation in medical AI, with a focus on achieving better
clinical decisions, advancing personalized medicine, and streamlining healthcare efficiency.
Education
Stanford University, Stanford, CA
Ph.D. in Electrical Engineering
2012 - 2015
University of California - Los Angeles, Los Angeles, CA
Master of Science in Electrical Engineering
2009 - 2012
Zhejiang University, Hangzhou, China
Bachelor of Engineering in Information and Electronic Engineering
2005 - 2009
Professional Experience
Principal Machine Learning Scientist
Stanford University, Stanford, CA
2023 - present
Data Science Lead
LVIS Corporation, Palo Alto, CA
2015 - 2023
Awards
Editor's Pick in the Magnetic Resonance in Medicine
2018 Nov
Best Overall Poster, NVIDIA GPU Technology Conference
2018
Best Healthcare Poster, NVIDIA GPU Technology Conference
2018
Magna Cum Laude Merit Award, The International Society of Magnetic Resonance and Medicine (ISMRM)
2013
Zhejiang Province Outstanding Undergraduate Award
2009
Best Undergraduate Thesis Award, Zhejiang University
2009
Journal Publications
Diagnostic accuracy of quantitative multicontrast 5-minute knee MRI using prospective artificial intelligence image quality enhancement
American Journal of Roentgenology, 216(6), pp.1614-1625. 2021.
Utility of deep learning super-resolution in the context of osteoarthritis MRI biomarkers
Journal of Magnetic Resonance Imaging, 51(3), pp.768-779. 2020.
Super-resolution musculoskeletal MRI using deep learning
Top 20 Most Downloaded Papers 2018-2019 of Magnetic Resonance in Medicine, And Editor's Pick of 2018 Nov
Magnetic resonance in medicine, 80(5), pp.2139-2154. 2018.
Comparison of fMRI analysis methods for heterogeneous BOLD responses in block design studies
NeuroImage. doi: 10.1016/j.neuroimage.2016.12.045. 2016.
High spatial resolution compressed sensing (HSPARSE) functional magnetic resonance imaging
Magnetic Resonance in Medicine. doi:10.1002/mrm.25854. 2016.
Combining optogenetic stimulation and fMRI to validate a multivariate dynamical systems model for estimating causal brain interactions
NeuroImage. 132:398-405. 2016.
Frequency-selective control of cortical and subcortical networks by central thalamus
eLife 4 e09215. 2015.
Optogenetic fMRI reveals distinct, frequency-dependent networks recruited by dorsal and intermediate hippocampus stimulations
NeuroImage. 107:229-241. 2015.
Patents
Efficacy and/or treatment parameter recommendation using individual patient data and therapeutic brain network maps
US 2019/0142338 A1, 2019.
Systems and methods for generating thin image slices from thick image slices
US Patent Appl. 16/979,104, 2018.
Conference Publications
Convolutional neural network for real-time high spatial resolution functional magnetic resonance imaging
Intl Soc Magn Reson Med, Montreal, 2019..
Evaluating the Use of Deep learning Super-Resolution for Obtaining Osteoarthritis Biomarkers
Intl Soc Magn Reson Med, Montreal, 2019.
Accurate T2 relaxometry with simultaneous high-resolution structural imaging using deep learning
Intl Soc Magn Reson Med, Montreal, 2019.
Super-resolution MRI using deep learning
Intl Soc Magn Reson Med, Paris, 2018.
Deep learning super-resolution enables rapid simultaneous morphological and quantitative magnetic resonance imaging
Medical Image Computing and Computer Assisted Intervention Machine Learning for Medical Image Reconstruction (pp. 3-11). Springer, Cham. (2018) pre-print: arXiv:1808.04447.
Automated knee cartilage segmentation with very limited training data: combining convolutional neural networks with transfer learning
Intl Soc Magn Reson Med, Paris, 2018.
Enhancing MRI resolution and fully-automating tissue segmentation using deep learning
Best Poster Award, NVIDIA GTC
NVIDIA GPU Technology Conference, San Jose, CA. 2018.
HSPARSE - a compressed sensing based high spatial resolution fMRI method
Society for Neuroscience 2015 annual meeting, Chicago, IL, USA, 449.16.
Dynamic control of forebrain by central thalamus
Society for Neuroscience 2015 annual meeting, Chicago, IL, USA, 449.20.
Comparison of fMRI analysis methods for accurate detection of heterogeneous hemodynamic responses
Society for Neuroscience 2015 annual meeting, Chicago, IL, USA, 449.13.
Optimized compressed sensing reconstruction with parallel computation for high spatial resolution functional magnetic resonance imaging
Society for Neuroscience 2014 annual meeting, Washington D.C., USA, 184.10.
Whole brain dissection of central thalamic circuit function with optogenetic fMRI
Society for Neuroscience 2014 annual meeting, Washington D.C., USA, 851.10.
Optogenetic fMRI reveals distinct, frequency-dependent networks recruited by dorsal and intermediate hippocampus stimulations
Society for Neuroscience 2014 annual meeting, Washington D.C., USA, 851.11.
GPU based fast inverse Gauss-Newton motion correction method for high throughput ofMRI
Magna Cum Laude Merit Award
Proc. Intl. Soc. Mag. Reson. Med 21st annual meeting, Salt Lake City, UT, USA, 2013, p4420.
On-demand generation of seizures with defined network propagation pathways
Society for Neuroscience 2013 annual meeting, San Diego, CA, USA, 336.21.
Brain circuit analysis with real-time optogenetic functional magnetic resonance imaging (rt-ofMRI)
Proc. Intl. Soc. Mag. Reson. Med 20th annual meeting, Melbourne, Australia, 2012; p4604.
Compressed sensing enabled ultra-high resolution optogenetic functional MRI
Proc. Intl. Soc. Mag. Reson. Med 20th annual meeting, Melbourne, Australia, 2012; p2051.
Real-time optogenetic functional magnetic resonance imaging (rt-ofMRI) using graphic processing unit (GPU) based parallel computation
Society of neuroscience 2011 annual meeting, Washington D.C., USA, 114.05.
Services
Reviewer, IEEE Journal of Biomedical and Health Informatics
2021-Now
Reviewer, ETRI Journal
2018
Reviewer, SPIE Journal of Medical Imaging
2017-Now
Reviewer, Neuroimage
2016