Session Abstract – PMWC 2022 Silicon Valley


Track Chairs:
Sharmila Majumdar, UCSF

The use of Artificial Intelligence (AI) in diagnostic medical imaging is undergoing extensive evaluation. AI has shown impressive accuracy and sensitivity in the identification of imaging abnormalities and promises to enhance tissue-based detection and characterization. This track will explore technical advancements in clinical machine learning and use cases in radiology.

Sessions:

  • Fireside Chat: Regulatory Considerations for AI & Machine Learning in Medical Imaging
    - Bakul Patel, FDA
    - Sean Khozin, CancerLinQ
  • AI Roadmap: Opportunities and Challenges (PANEL)
    Session Chair: Sean Khozin, CancerLinQ
    - Patrick Loerch, Gilead
    - Daniel J. Arbess, Xerion Investments LLC
    - Nigam Shah, Stanford
  • Technical Advancements in Clinical Machine Learning - What's New in 2021 (PANEL)
    Session Chair: Sharmila Majumdar, UCSF
    - Jayashree Kalpathy-Cramer, Harvard
    - John Mongan, UCSF
  • Image Applications for Clinical Diagnosis
    Session Chair: Esther L. Yuh, UCSF
    - Thorsten Fleiter, U. of Maryland
    - Imon Banerjee, Mayo Clinic
    - Anant Madabhushi, Case Western Reserve University
    - Pratik Mukherjee, UCSF
    - Lawrence Schwartz, Columbia University Medical Center
  • The Impact of AI and High-Resolution Digital Slide Imaging
    - Christopher Coley, Epredia
  • Evaluation of AI Software for Radiological Applications (PANEL)
    Session Chair: Kyung Hyun Sung, UCLA
    - Akshay Chaudhari, Stanford
    - Mona Flores, NVIDIA
  • PMWC 2022 Showcase (NCI)
    - Eric Horler, AIQ Solutions

  •  Session Chair Profile

    Ph.D., Professor and Vice Chair, UCSF

    Biography
    Sharmila Majumdar, Ph.D., obtained her Ph.D. degree from Yale University in Engineering and Applied Sciences. After a short stay at Yale she joined UCSF as an Assistant Professor in 1989. Her research on machine and deep learning has a focus on imaging; prior work has focused on imaging technology development and translational, development of image processing, computer vision has been focused in the areas of osteoporosis, osteoarthritis, and lower back pain. She is the Executive and Scientific Director of the Center for Intelligent Imaging at UCSF. She was selected as a fellow of the American Institute of Medical and Biological Engineers in 2004, a fellow of the International Society of Magnetic Resonance in Medicine (ISMRM) in 2008, and awarded the ISMRM Gold medal in 2016. She is an active participant in the PMWC meetings. Her work over the last three decades has been extensively supported by the NIH and industry.


     Session Chair Profile

    M.D., M.P.H., CEO, CancerLinQ

    Biography
    Dr. Khozin is an oncologist and physician-scientist, currently serving as Chief Executive Officer of CancerLinQ, a company focused on advancing cancer care quality and research. Previously, he was Global Head of Data Strategy and Data Science Innovation at J&J, leading a worldwide multidisciplinary team charged with the design and implementation of pioneering data science solutions to support the development of innovative medicines. Before joining J&J, Dr. Khozin led FDA Oncology Center of Excellence’s bioinformatics, regulatory science, and clinical trial innovation efforts and was the founding Executive Director of Information Exchange and Data Transformation (INFORMED), the FDA’s first data science and technology incubator that played a pivotal role in the agency’s technology modernization efforts. Before his tenure in federal government, Dr. Khozin was the cofounder of Hello Health, a company focused on developing integrated telemedicine, point-of-care data visualization, and advanced analytical systems for optimizing patient care and clinical research.

    Talk
    AI Roadmap: Opportunities and Challenges
    This session outlines current challenges and opportunities related to the use of AI to advance therapeutic development and care delivery, with a special focus on medical imaging.


     Session Chair Profile

    M.D., Ph.D., Professor, UCSF

    Biography
    Dr. Yuh is a neuroradiologist and physicist who interprets images of the brain, neck and spine as attending physician at UCSF and at Zuckerberg San Francisco General Hospital. She works with UC Berkeley computer scientists and UCSF radiologists on machine learning algorithm applications in traumatic brain injury. She ed FDA approval in 2019 of the first-ever imaging biomarker to select patients for traumatic brain injury therapeutic clinical trials, under the FDA’s Medical Device Development Tool (MDDT) program. A graduate of Stanford University School of Medicine, she completed training in radiology and neuroradiology at UCSF in 2009. She has been nominated for the UCSF Exceptional Physician Award, and is a prior Outstanding Teaching Fellow of the UCSF Department of Radiology and Biomedical Imaging.


     Session Chair Profile

    Ph.D., Associate Professor, UCLA

    Biography
    Dr. Sung’s primary research focuses on the development of novel medical imaging methods and artificial intelligence using magnetic resonance imaging (MRI). He received a Ph.D. degree in Electrical Engineering from the University of Southern California, Los Angeles, in 2008, and from 2008 to 2012, he finished his postdoctoral training at Stanford in the Departments of Radiology. He joined the University of California, Los Angeles (UCLA) Department of Radiological Sciences, in 2012. His research interest is to develop fast and reliable MRI techniques that can provide improved diagnostic contrast and useful information. In particular, his research group (https://mrrl.ucla.edu/sunglab/) is currently focused on developing advanced deep learning algorithms and quantitative MRI techniques for early diagnosis, treatment guidance, and therapeutic response assessment for oncologic applications. Such developments can offer more robust and reproducible measures of biologic markers associated with human cancers.


     Speaker Profile

    MBA, Associate Director for Digital Health, Center for Devices and Radiological Health, FDA

    Biography
    Bakul Patel is Associate Director for Digital Health, at the Center for Devices and Radiological Health (CDRH), at the Food and Drug Administration (FDA). Mr. Patel leads regulatory policy and scientific efforts at the Center in areas related to emerging and converging areas of medical devices, wireless and information technology. This includes responsibilities for mobile health, health information technology, cyber security, medical device interoperability, and medical device software. Mr. Patel is the FDA liaison between the Federal Communications Commission (FCC) and the Office of the National Coordinator (ONC). Since its inception in 2013, Bakul chairsthe International Medical Device Regulators Forum (IMDRF) “software as a medicaldevice” working group, a global harmonization effort. Before joining FDA, Mr. Patel held key leadership positions working in the telecommunications industry, semiconductor capital equipment industry, wireless industry and information technology industry. His experience includes Lean Six Sigma, creating long and short‐term strategy, influencing organizational change, modernizing government systems, and delivering high technology products and services in fastpaced, technology‐intensive organizations. Mr. Patel earned an MS in Electronic Systems Engineering from the University of Regina, Canada, and an MBA in International Business from The Johns Hopkins University.


     Speaker Profile

    Ph.D., SVP Clinical Data Sciences, Gilead

    Biography
    Patrick Loerch has over 20 years of experience in data sciences, real world evidence and genomics research spanning all phases of R&D within the pharmaceutical industry. Dr Loerch currently serves as Senior Vice President, Clinical Data Sciences at Gilead Sciences. In addition to oversight of the Biometrics, Clinical Bioinformatics and Real World Evidence organizations, Dr. Loerch is accountable for the build out and integration of Enterprise Data Sciences capabilities to accelerate the discovery, development and delivery of new medicines. Dr. Loerch’s prior roles include leadership positions at Johnson & Johnson, Celgene and Merck. He began his career at Rosetta Inpharmatics, a genomics start-up in Seattle, WA that was later acquired by Merck. Dr. Loerch received his PhD in Biostatistics from the Harvard University and his BS in Biochemistry from Washington State University.


     Speaker Profile

    CEO, Xerion Precision Bio

    Biography
    Investor & patients' advocate in systems biology, precision neuroscience. Served on steering committee of the MIT/UCSB Alzheimer’s X initiative, and is on the financing structure committee of the American Medical Association/One Mind Healthy Brain Global Initiative; the Board of Directors of the Global Virus Initiative and the corporate advisory Board of Cancer Expert Now. Adviser to CEO of CancerLinq. Dan is a multi-asset class investor, lawyer and social entrepreneur whose 35-year career has been defined by purposeful engagement as an adviser and investor in important global developments, most recently through his $3.5 Billion Xerion Hedge Funds.


     Speaker Profile

    Ph.D., Scientific Director, CCDS, Harvard

    Biography
    Jayashree Kalpathy-Cramer is the Director of the QTIM lab and the Center for Machine Learning at the Athinoula A. Martinos Center for Biomedical Imaging and an Associate Professor of Radiology at MGH/Harvard Medical School. Dr. Kalpathy-Cramer is also Scientific Director at MGH & BWH Center for Clinical Data Science. An electrical engineer by training, she worked in the semiconductor industry for a number of years. After returning to academia, she is now focused on the applications of machine learning and modeling in healthcare. Her research interests include medical image analysis, machine learning and artificial intelligence for applications in radiology, oncology and ophthalmology. The work in her lab spans the spectrum from novel algorithm development to clinical deployment. She is passionate about the potential that these techniques have to improve access to healthcare in the US and worldwide. Dr. Kalpathy-Cramer has authored over 100 peer-reviewed publications and has written over 10 book chapters.


     Speaker Profile

    M.D., Chairman, Columbia University Medical Center

    Biography
    Lawrence H. Schwartz, M.D. is the James Picker Professor and Chairman of the Department of Radiology at Columbia University Medical Center and is also Radiologist-in-Chief at New York-Presbyterian Hospital/Columbia University Medical Center. In his capacity as Chairman, Dr. Schwartz oversees a 75 clinical member department and directs its patient care, research and educational initiatives. Before joining Columbia, Dr. Schwartz was on the faculty at Memorial Sloan-Kettering Cancer Center for approximately 18 years, most recently as the Vice Chair, Technology Development in the Department of Radiology as well as Director of MRI. Dr. Schwartz’s academic interests are in the development of novel imaging biomarkers in oncology, both clinical care and drug discovery. He is an active member of the Quantitative Imaging Biomarker Alliance and the Oncology Biomarker Qualification Initiative.


     Speaker Profile

    M.D., Professor, University of Maryland School of Medicine

    Biography
    Thorsten Fleiter is a member of the radiology team at the Nation’s first and only integrated Trauma Hospital concentrating on the development of new imaging methods applying AI and robotic techniques to create autonomous systems to shorten the time lost for the detection of life-threatening injuries in severe trauma. The goal of his group is to create less user independent solutions specifically for pre-hospital and remote scenarios to ultimately improve the survival of trauma victims. He led the development of novel imaging methods for ultrasound and computed tomography primarily for emergency imaging and developed multiple advanced acquisition and visualization methods including 4D-ultassound, virtual bronchoscopy/colonoscopy as well as new tools for precision surgery guidance using novel 3D printing methods for spine, facial bone, and dental applications. His group earned several awards for the advanced visualization and simulation of dynamic processes like blood- and airflow to predict pathophysiological changes of aneurysms and stents.

    Talk
    Applying AI techniques in Severe Trauma Imaging
    Routine sub-millimeter scans of the entire body of trauma victims are typically producing 3000-5000 images that must be reviewed instantaneously to detect life threatening injuries. New AI image analysis tools and workflow alterations to further improve this process and shorten the turnaround times will be discussed in this presentation.


     Speaker Profile

    Ph.D., Senior Associate Consultant, Lead AI Scientist, Associate Professor, Mayo Clinic

    Biography
    I am an Associate consultant and lead AI scientist in Mayo Clinic Arizona. I am affiliated within ASU School of Computing and Augmented Intelligence. Before joining Mayo clinic, I was working as an Assistant Professor in Emory University with joint affiliation in Georgia Tech. I did my postdoc training from Biomedical data science department at Stanford University and completed my doctoral degree in Computer Science as a Marie Curie fellow from National Council of Research, Italy. My current research is focused on unstructured medical data analysis and integration of multisource medical data from varying hospital systems for building predictive model to benefit cancer diagnosis and treatment. I successfully lead multiple federal and non-federal research projects related to machine and deep learning based research projects. I am currently leading multiple innovative multi-institutional research projects related to cancer informatics which involves both academic (Emory, Duke, Stanford, Harvard, Intermountain, IU) and industrial (Philips, GE healthcare) partners.

    Talk
    Fusion of Multi-modal Data for Clinical Event Prediction
    Even though advancements in deep learning techniques carry the potential to make significant contributions to healthcare, most of current models consider only a single input data stream without incorporating data that can inform clinical context. Yet in practice, non-imaging data based on the clinical history and laboratory data enable physicians to interpret imaging findings in the appropriate clinical context, leading to a higher diagnostic accuracy. Models must also achieve the capability to process contextual clinical data in addition to pixel or other sensor data. This talk will present multiple fusion machine learning methodologies including graph-based models, with boosted performance by integrating the clinical context with the imaging data applied to different clinical context.


     Speaker Profile

    Ph.D., Professor, Case Western Reserve University

    Biography
    Dr. Anant Madabhushi is Director of the Center for Computational Imaging and Personalized Diagnostics (CCIPD) and Donnell Institute Professor, Department of Biomedical Engineering at Case Western Reserve University. He is a Research Health Scientist at the Louis Stokes, Cleveland Veterans Administration Medical Center. Dr. Madabhushi has authored over 400 peer-reviewed publications and over 100 patents issued or pending. He is a, Fellow of the American Institute of Medical and Biological Engineering (AIMBE), and the Institute for Electrical and Electronic Engineers (IEEE) and the National Academy of Inventors (NAI). His work on "Smart Imaging Computers for Identifying lung cancer patients who need chemotherapy" was called out by Prevention Magazine as one of the top 10 medical breakthroughs of 2018. In 2019, Nature Magazine hailed him as one of 5 scientists developing "offbeat and innovative approaches for cancer research". Dr. Madabhushi was named to The Pathologist’s Power List in 2019, 2020 and 2021.

    Talk
    AI for Radiology and Pathology for Precision Medicine
    Our group has been pioneering computerized feature analysis methods for extracting subvisual attributes for characterizing disease appearance and behavior on radiographic (radiomics) and digitized pathology images (pathomics). In this talk I will discuss how these radiomic and pathomic approaches can be applied to predicting disease outcome, recurrence, progression and response to therapy in the context of prostate, brain, rectal, oropharyngeal, and lung cancers. Additionally I will also discuss some recent work on looking at use of pathomics in the context of racial health disparity and creation of more precise and tailored prognostic and treatment response prediction models.


     Speaker Profile

    M.D., Ph.D., Professor, UCSF

    Biography
    Dr. Mukherjee is a clinical neuroradiologist whose research involves technical development and basic and clinical neuroscience applications of methods for mapping tissue microstructure, connectivity and function in the human brain. Recent work includes diffusion MRI, resting state fMRI and magnetoencephalography (MEG) to study cerebral connectivity, including the whole-brain macroscale network known as the “connectome”, in human brain development, neurodevelopmental disorders and traumatic brain injury (TBI). He has special experience in standardizing structural MRI, diffusion MRI, and fMRI for large-scale multi-site projects and managing the resulting Big Data using cutting-edge informatics platforms and machine learning analytics, which is essential for the clinical translation of new imaging technology. He is a PI and Imaging Core leader of the Transforming Research and Clinical Knowledge in TBI (TRACK-TBI) US nationwide study and also a PI and Clinical Team Leader of two NIH Brain Research through Advancing Innovative Neuro-technologies (BRAIN) grants to produce transformational advances in ultra-high field brain MR imaging technology.

    Talk
    Artificial Neural Networks for Understanding Human Brains
    Artificial neural networks (ANNs), including deep learning, were originally inspired by biological neural networks. In turn, ANNs are now being used to study the human brain. We show that deep learning of fMRI can map the brain’s hierarchical functional organization by detecting “meta-networks” that represent interactions of low-order intrinsic connectivity networks.


     Speaker Profile

    Ph.D., Assistant Professor, Stanford University

    Biography
    Dr. Chaudhari is an Assistant Professor in the Department of Radiology and (by courtesy) Biomedical Data Science. He is also the Associate Director of Research and Education at the Stanford AIMI Center. His primary research interests lie at the intersection of artificial intelligence and medical imaging. Dr. Chaudhari graduated from UCSD with a B.S. in Bioengineering in 2012. He completed his Ph.D. from Stanford University’s Department of Bioengineering in 2017, focusing on novel MRI methods for musculoskeletal imaging; supported through the National Science Foundation Graduate Research Fellowship, the Whitaker Fellowship, and the Siebel Fellowship. Dr. Chaudhari trained as a postdoctoral fellow in Radiology at Stanford University, where he combined machine learning with medical imaging acquisition and analysis. Dr. Chaudhari has won many awards, including the W.S. Moore Young Investigator Award, the Junior Fellow Award, and an Outstanding Teacher Award from the International Society for Magnetic Resonance in Medicine.


     Speaker Profile

    M.D., Global Head of Medical AI, NVIDIA

    Biography
    Dr. Flores oversees NVIDIA’s AI initiatives in medicine and healthcare to bridge the chasm between technology and medicine. Dr. Flores first joined NVIDIA in 2018 with a focus on developing the healthcare ecosystem. Before joining NVIDIA, she served as the chief medical officer of digital health company Human- Resolution Technologies after a 25 + year career in medicine and cardiothoracic surgery. She received her medical degree from Oregon Health and Science University followed by a general surgery residency at the University of California at San Diego, a Postdoctoral Fellowship at Stanford, and a cardiothoracic surgery residency and fellowship at Columbia University in New York. Dr. Flores also has a Master of Biology from San Jose State and early in her career received an MBA from the University at Albany School of Business. She initially worked in investment banking for a few years before pursuing her passion for medicine and technology.


     Speaker Profile

    Global Marketing Development Manager, Digital Pathology, Epredia

    Biography
    Chris Coley, PA/HT(ASCP), has an extensive Anatomical Pathology background followed by various roles in sales and marketing for comprehensive product lines in the both clinical and research segments for over 25 years. His experiences enable him to implement market enhancing technologies that professionally align with Epredia’s mission and vision to improve lives by enhancing precision cancer diagnostics.

    Talk
    The Impact of Artificial Intelligence and High Resolution Digital Slide Imaging on Precision Medicine
    Epredia hosts a partnership panel discussion on how 3DHISTECH acquired digital slide images are used with AI solutions from Aiforia and PaigeAI to drive accuracy and efficiency in the study and diagnosis of disease.


     Speaker Profile

    M.D., Ph.D., Assoc Prof, Assoc Chair Translational Informatics, Director Ci2, UCSF

    Biography
    Dr. Mongan's research focuses on artificial intelligence in medical imaging. He was the senior author and primary investigator on a project that developed artificial intelligence for the detection of pneumothorax (collapsed lung); in partnership with General Electric, the algorithm developed in this project achieved FDA clearance and is currently commercially available on portable X-ray machines. He is the lead author on the Checklist for Artificial Intelligence in Medical Imaging (CLAIM), a guideline used by several journals to promote reproducibility in artificial intelligence publications, and is the lead author on a publication drawing lessons for the safe implementation of artificial intelligence in medicine from the 737 Max disasters. He chairs the Machine Learning Steering Committee of the Radiological Society of North America (RSNA, the world’s largest radiology specialty society) and serves on the editorial board of the journal Radiology: Artificial Intelligence.


     Speaker Profile

    M.D., Ph.D., Professor of Medicine, Stanford University; Associate CIO for Data Science, Stanford Healthcare

    Biography
    Dr. Nigam Shah is Professor of Medicine (Biomedical Informatics) at Stanford University, and serves as the Associate CIO for Data Science for Stanford Health Care. Dr. Shah's research focuses on bringing machine learning to clinical use safely, ethically and cost-effectively. Dr. Shah was elected into the American College of Medical Informatics (ACMI) in 2015 and was inducted into the American Society for Clinical Investigation (ASCI) in 2016. He holds an MBBS from Baroda Medical College, India, a PhD from Penn State University and completed postdoctoral training at Stanford University.