Topic 1:

AI to improve accuracy of diagnosis and health risk assessment

Oral Presenters

Krzysztof Geras, PhD, MSc

Assistant Professor, Department of Radiology, New York University School of Medicine

"Deep Neural Networks Improve Radiologists' Performance in Breast Cancer Screening"

Krzysztof is an assistant professor at NYU School of Medicine and an affiliated faculty at NYU Center for Data Science. His main interests are in unsupervised learning with neural networks, model compression, transfer learning, evaluation of machine learning models and applications of these techniques to medical imaging.



He previously did a postdoc at NYU with Kyunghyun Cho, a PhD at the University of Edinburgh with Charles Sutton and an MSc as a visiting student at the University of Edinburgh with Amos Storkey. His BSc is from the University of Warsaw. He also did industrial internships in Microsoft Research (Redmond, working with Rich Caruana and Abdel-rahman Mohamed), Amazon (Berlin, Ralf Herbrich's group), Microsoft (Bellevue) and J.P. Morgan (London).


Yan Wu, PhD

Instructor, Radiation Oncology, Stanford University

"Deciphering Quantitative Tissue Parameters from Single Qualitative MR Images by Deep Learning"

Dr. Yan Wu is an instructor in the Department of Radiation Oncology of Stanford University. She conducts research on deep learning empowered MR imaging at Laboratory of Artificial Intelligence in Medicine and Biomedical Physics. She has developed several frameworks for generic MRI reconstruction and a quantitative MRI approach that derives multi-parametric maps from a single qualitative image.


Previously Dr. Wu got her PhD from University of Wisconsin Madison with research focused on accelerated dynamic MR imaging methods. Then she worked on perfusion MRI and machine learning models for radiomics at Robert Wood Johnson Medical School.


Topic 2:

AI to improve selection of treatment options

Oral Presenters

Peter McCaffrey, MD

CTO & Co-Founder, VastBiome

"ImmuNET: A Neural Network to Improve the Selection of Immunotherapy in Cancer"

Peter McCaffrey is a board-certified Pathologist as well as Co-Founder and CTO of VastBiome, a drug discovery company mining the human gut microbiome for novel cancer therapeutics. Peter completed his medical training at The Johns Hopkins University School of Medicine and his Residency in Pathology at The Massachusetts General Hospital where he also served as Chief Resident.


After Residency, he completed a Fellowship in Biodesign at the Texas Medical Center in Houston where VastBiome originated. Peter's interests in microbiology stem from his time as a Research Fellow at the KwaZulu-Natal Research Institute for Tuberculosis and HIV in Durban, South Africa where his work focused on applying computational biology to define personalized antibiotic resistance profiles based upon bacterial and viral sequence data for individual patients. At VastBIome, Peter applies these skills to map the mechanisms through which gut microbes control the host immune system in the setting of cancer and immunotherapy.


Marta Skreta, B.Sc.

Research Student in the Centre for Computational Medicine, The Hospital for Sick Children

"Automated prognosis of prenatal hydronephrosis"

Marta Skreta is Master’s student in Computer Science at the University of Toronto. Her research focuses on developing and applying machine learning algorithms to language and image data in healthcare. Marta has become involved in the development of clinical note-taking and decision support systems at the Hospital for Sick Children in Toronto under the supervision of Dr. Michael Brudno. Her goal is to improve the translation and usability of predictive models in clinical settings.


Marta is the recipient of the Vector Scholarship in Artificial Intelligence and the Canada Graduate Scholarships—Master’s Award. Previously, she completed a BSc in Chemical Biology at McMaster University under the Schulich Leader Scholarship. Marta is also a founding member of HER CODE CAMP, a camp that introduces high school girls in Toronto to coding.


Panel Participants

Geoffrey Schau, MS

PhD Candidate, Oregon Health & Science University

"Deep Neural Estimation of Metastatic Origin of Liver Cancer"

Geoffrey Schau is currently a PhD candidate in the Biomedical Engineering Department and Computational Biology Program at Oregon Health & Science University. He is working on projects focusing on the design and application of deep learning systems to improve our understanding of the spatial role of cellular organization in cancer through the OHSU Center for Spatial Systems Biomedicine under the mentorship of Dr. Young Hwan Chang.


Prior to beginning his PhD, he worked for three years as an applied research engineer in the medical device industry developing computational algorithms for intracardiac signal detection and processing. He earned an undergraduate degree in biomedical engineering from Rose-Hulman Institute of Technology and a master’s degree in electrical engineering from Portland State University before coming to OHSU.


Topic 3:

AI to improve step-by-step clinical pathways used to apply treatments

Oral Presenters

Alex Deakyne, MS

Professional Researcher at University of Minnesota, Department of Surgery, Visible Heart Lab

"Deep Learning pipeline for the automatic segmentation and 3D model generation of the four heart chambers; applied uses in pre-surgical planning"

Alex Deakyne is a research scientist and PhD student in the Department of Surgery at the University of Minnesota. He is pursuing his degree in Bioinformatics and Computational Biology with a focus on deep learning and virtual reality for medical imaging. He has an MS in Bioinformatics and Computational Biology from the University of Minnesota and a BS in Mechanical Engineering from Iowa State University.

Tony Duan, MS

 AI Resident, Microsoft

"Guiding Heparin Treatment with Reinforcement Learning"

Tony is currently an AI Resident at Microsoft. He recently graduated from the MS program in CS at Stanford, where he worked on various applications of machine learning to health care. Areas of interest include treatment effect estimation, survival predictions, and medical imaging.

Panel Participants

Erik Burlingame, MSc

Computational Biology Student, Oregon Health & Science University

"SHIFT: deep learning-based inference of biomarker distribution in histopathology images with speedy histopathological-to-immunofluorescent translation"

Erik Burlingame is a PhD candidate under the mentorship of Dr. Young Hwan Chang in the Computational Biology Program and Biomedical Engineering Department at Oregon Health and Science University. His research focuses on deep learning applications for medical imaging, in particular, the application of generative models to learn relationships between imaging modalities common to digital pathology. He aims to build automated systems that simultaneously economize and democratize advanced imaging technologies.

Sidra Xu

Student, The Harker School

"Machine Learning-Assisted Prediction of Surgical Mortality of Lung Cancer Patients"

Sidra Xu is a junior at The Harker School and a research intern at Stanford Medical School in Dr. Sean Wu's lab. Ever since she was introduced to machine learning, she has been captivated by the plethora of applications of artificial intelligence to various fields, especially medicine.


Eager to bridge the two disciplines, Sidra has conducted research in bioinformatics for three years and presented her work at many science fairs and conferences, including ICDM this July. In the future, she hopes to harness the power of AI to improve healthcare quality as both a doctor and researcher. Sidra is also a founder and president of ETForum, an online community for youths to discuss issues on technology and ethics, vice president of GPL, an organization dedicated to empowering girls in learning programming, and director of Opportunity X, a nonprofit bringing science research opportunities to underrepresented students.


Topic 4:

AI to detect and correct failures in clinician, patient and lay care-giver treatment actions inside and outside of healthcare facilities

Oral Presenters

J Peter Campbell, MD, MPH

Assistant Professor, Casey Eye Institute, Oregon Health & Science University

"Reducing childhood blindness from retinopathy of prematurity using artificial intelligence"

Dr. Campbell received his MD and MPH degrees from Johns Hopkins and completed a fellowship in vitreoretinal surgery at the Oregon Health & Science University where he is now an Assistant Professor.  He has an active vitreoretinal practice and is one of the only pediatric retinal specialists in the Pacific Northwest.


He is currently on an NIH K12 training grant focused on development objective tools for quantification of disease severity in retinopathy of prematurity (ROP), and is a recipient of a Career Development Award from the Research to Prevent Blindness. His academic expertise involves the use of artificial intelligence (AI) and imaging devices for diagnosis of ROP. Dr. Campbell has published more than 60 peer-reviewed publications, and is one of the founding members of the American Academy of Ophthalmology AI Task force, as well as a member of the third International Classification of Retinopathy of Prematurity Committee. 



Maksim Tsvetovat, PhD

Founder & CTO, Open Health Network

"A Conversational Dialogue Agent to Improve Outcomes in Healthcare"

Dr. Maksim Tsvetovat is a co-founder of Open Health Network, an AI-powered patient experience management platform. He is a Visiting Faculty Member at the George Washington University's School of Engineering and Applied Sciences, author of over 50 peer reviewed publications and a textbook in the fields of data science, artificial intelligence and simulation modeling.


Dr. Tsvetovat received his Ph.D. from Carnegie Mellon University's School of Computer Science and was previously on faculty at George Mason University. In his free time, such that it is, Dr. Tsvetovat races Ironman triathlons and plays in a rock band.


Topic 5:

Innovation in AI methods that increase AI’s capacity to improve healthcare

Oral Presenters

Gabriel Sanchez, PhD

Co-Founder & CEO, Enspectra Health

"Noninvasive cellular imaging in live skin makes histopathology accessible for AI innovation"

Gabriel Sanchez is the CEO and co-founder of Enspectra Health, a Stanford University spin out currently in residence at the Fogarty Institute for Innovation in Mountain View CA. Enspectra has created a portable microscope that images real-time cellular anatomy in living tissue without cuts or stains.


This technology introduces a new field of medical imaging, non-invasive histopathology, and the feature rich images are ideal for pairing with emerging AI tools in machine vision. Enspectra’s mission is to bring real-time, digital histopathology into mainstream Dermatology. Gabriel received his B.S. in mechanical engineering from MIT, and an M.S. and Ph.D. in mechanical engineering from Stanford University.


Vincent Tseng

Ph.D. Candidate, Information Science, Cornell University

"Developing Clinically Interpretable Machine Learning Models to Predict Fine-Grained Symptom Trajectory of Schizophrenia and Identify Patients At Risk"

Vincent Tseng is a Ph.D. candidate in Information Science at Cornell University. His research focuses on developing mobile sensing techniques and digital markers to help assess people’s cognitive performance, such as alertness and attention, and mental well-being. He is also interested in designing intervention technologies that help improve people’s cognitive performance.

Panel Participants

Wisdom d'Almeida, MTech

Research Intern, Mila - Quebec Artificial Intelligence Institute

"Optimizing Radiology Report Generation for Clinical Pertinence"

Wisdom is a research intern at MILA, working with Yoshua Bengio and Alain Tapp. He holds a master's degree from KIIT in India and a B.S. from Université de Lomé in Togo where he grew up. Wisdom's current research gravitates around Grounded Language Learning and AI explainability. In the past, he worked on natural language understanding for common-sense reasoning, with application to areas such as healthcare.


His master's dissertation was about medical report generation with natural language explanations. Wisdom’s works in AI won a Government of India National Award in 2018 and he previously interned at Google in San Francisco, where he launched Google Cloud support for emerging languages.


Zhitao Li, PhD

Postdoctoral Research Fellow, Radiology, Stanford

"Estimating Systematic Imperfections and Performing Data-Driven Reconstruction with Deep Neural Networks for Non-Cartesian Wave-Encoded Magnetic Resonance Imaging"

Zhitao Li is a postdoc at the Stanford School of Medicine doing MRI imaging acqusition and reconstruction work with machine learning techniques. Zhitao joined Stanford after obtaining his PhD from The University of Arizona in electrical and computer engineering (2019). His research now focuses on the development of Non-Cartesian MRI techniques.

Edgar Anselmo Rios Piedra, PhD

Postdoctoral Research Fellow, Radiology, Stanford

"DeepKidney: Deep segmentation of MR images for automated glomerular function quantification in heterogeneous pediatric patients"

Edgar Rios Piedra is a postdoctoral fellow at the Stanford School of Medicine doing medical image analysis and machine learning research. He joined Stanford after obtaining his PhD from UCLA in medical image analysis, machine learning and pattern recognition (2018). His research now centers on the development of methods for computer-aided diagnosis systems, particularly applied to magnetic resonance imaging (MRI) on pediatric patients with diverse pathologies.

Topic 6:

AI to improve patients’ ability to self-assess symptoms, select and implement self-care options to avoid use of or more successfully partner with health care professionals

Oral Presenters

Michael Lu, MD, MPH

Director of Research, Division of Cardiovascular Imaging, Massachusetts General Hospital & Harvard Medical School

"Deep learning to assess long-term mortality and phenotypic age from chest radiographs"

Michael T. Lu, MD, MPH, is the Director of Research, Cardiovascular Imaging at Massachusetts General Hospital (MGH) and Assistant Professor of Radiology at Harvard Medical School. 

As a practicing radiologist, Dr. Lu’s goal is to apply medical imaging to improve health. His research focuses on A) machine learning to predict health outcomes from imaging and B) clinical trials of cardiac CT. He co-chairs the Mechanistic Substudy of REPRIEVE, a multicenter randomized controlled trial of statins to reduce coronary plaque and prevent heart disease in persons living with HIV. 


Dr. Lu completed his BA, MD, and MPH degrees at Harvard University. His training includes a residency in Diagnostic Radiology at the University of California, San Francisco and fellowships in Thoracic and Cardiac Imaging at MGH.


Christine Meinders, MFA, MA

Founder, Feminist.AI

"Contextual Normalcy: a participatory artificial intelligence research project on creating community-driven mental health classifications and disagnoses"

Christine Meinders is a cultural AI designer and researcher who uses collaborative and inclusive design approaches to creating AI design tools and developing community-driven, social AI projects. She founded the community AI research group, Feminist.AI. She is currently an adjunct Music Faculty at CalArts and adjunct IXD Faculty at California College of the Arts, she holds an MFA in Media Design Practices from ArtCenter College of Design and an MA in Clinical Psychology from Pepperdine University.

Panel Participants

Albert Buchard, MD, MSc

Research Scientist, Babylon Health

"Tuning semantic consistency of active medical diagnosis: a walk on the semantic simplex"

Albert Buchard is a Research Scientist at Babylon Health's AI Research lab, where he works on automated and efficient decision making in healthcare. Fascinated by the potential of brain-computer interfaces early on, he honed his skills as a software engineer while studying for his Medical Degree at University Pierre and Marie Curie in Paris. Both an SNF and Ecole de L'Inserm MD-PhD Fellow, he worked across several fields of Neurosciences during his Master and Ph.D. work, as well as a clinician in Neurosurgery and Psychiatry.


Since he joined Babylon, he participated in the development of state of the art decision making algorithms for medical diagnosis and triage using model-based methods and deep reinforcement learning. In his spare time, he enjoys creating indie video-games and 3D art with his better half.


Saurabh Johri PhD

Chief Scientist, Babylon Health

"A comparative study of artificial intelligence and human doctors for the purpose of triage and diagnosis"

Saurabh is Chief Scientist at Babylon. He has been with Babylon since 2016 and in this time, has guided the team to develop Babylon's AI for the development of the triage, diagnostic and predictive models for healthcare, and applied the team’s research in Bayesian Machine Learning and Causal inference. Prior to Babylon, Saurabh spent time as a post-doctoral researcher at the MRC Centre for Outbreak Analysis & Modelling at Imperial College London.


During this time, he focused on the development of novel statistical machine learning methods to estimate poliovirus transmission from genetic sequence data, working in collaboration with colleagues at the CDC. Before his post-doctoral work, Saurabh completed his PhD in population genetics from Imperial College London, investigating the population genetics of Tuberculosis and predicting new drug targets from whole genome sequence data.



Esther Dyson

Topic 4: AI to detect and correct failures in clinician, patient and lay caregiver treatment actions inside and outside of healthcare facilities

Esther Dyson (@edyson on twitter) is executive founder of  Wellville (@WaytoWellville),  a ten-year, five-community  project dedicated to showing the value of investing in health (as opposed spending on care). Wellville advises the five US-based Wellville communities on accelerating their own health initiatives, in Clatsop County, OR; Lake County, CA; Muskegon County, MI; North Hartford, CT; and Spartanburg, SC. Dyson is the W2W lead for Muskegon, and is actively involved in  policy and fundraising for the project.  Wellville’s overall mission is to encourage society to think longer-term and more broadly - from self to community - by illustrating the social and financial benefits of collectively in human capital.


Aside from that full-time role, Dyson spends her extra time investing in and nurturing start- ups, with a recent focus on health care and some AI (somewhat constrained to avoid conflicts with Wellville). On the health side, she is an investor in 23andMe (also a director), 4D Healthware, Basil Health, Big Health, Boundless.ai, Care.Coach, CareMESH, Clover Health, Devoted Health, Doppel, Eligible, Enso Relief, Ezra.ai, Hawthorne Effect, HealthCelerate, HealthTap, i2Dx, Joany, MealShare, Medesk, MedicaSafe, mEquilibrium, Nanowear, NeuroGeneCES, Nuna, Omada Health, PatientsLikeMe/iCarbonX, PatientsKnowBest, Prognos.ai, Proofpilot, Resilient, Sapiens DS, Solera (also an advisor), StartupHealth, Supportiv, Syllable.ai, Tega Pharmaceutical, Tocagen, Trusty.care, Turbine.ai, Valkee, Virgo, X-VAX and Zipongo.


Matthew Lungren, MD, MPH

Topic 1: AI to improve accuracy of diagnosis and health risk assessment

Matthew Lungren MD MPH is the Associate Director of the Stanford Center for Artificial Intelligence in Medicine and Imaging and an Assistant Professor at Stanford University Medical Center. His leading research interests are in the field of machine learning in healthcare for diagnosis, prediction, and augmentation to improve healthcare outcomes. His work is regularly featured in national news outlets and regularly speaks on the topic of AI in healthcare for national meetings. 


Emily Melton

Topic 2: AI to improve selection of treatment options

Emily Melton is a co-founder and managing partner of Threshold Ventures, an early stage venture capital fund. Her current portfolio includes BetterUp, Elation Health, Imagen, Livongo, Ooda Health, Shift, Verge Genomics, Vineti, and Wellframe.  

Prior to co-founding Threshold Ventures, Emily was a managing partner at DFJ, where she sourced the firm’s investment in Box (NYSE:  BOX), and sourced and led the firm’s investments in Meebo (Google), Kudo (Google), Redfin (NASDAQ: RDFN), RichRelevance, and Flux (MTV Networks). She was also an investor and advisor to Pulse Network (LinkedIn), Notion, and Poshmark.


Emily holds a BA with honors and distinction in political philosophy and an MBA, both from Stanford University. 


Quentin Hardy

Topic 3: AI to improve step-by-step clinical pathways used to apply treatments

Quentin Hardy is the head of Editorial at Google Cloud. He writes about the ways that cloud computing technology, and the advent of computational intelligence worldwide, is reshaping society.

Before joining Google, Mr. Hardy was Deputy Technology Editor at The New York Times, where for many years he wrote about enterprise and large-scale computing. He has also been National Editor of Forbes Magazine, and a reporter with The Wall Street Journal in San Francisco and Tokyo. For 10 years he was part of “Forbes on Fox,” a weekend news program on Fox News.


His feature stories for The Journal have been widely anthologized, and his work on technology in Africa was given a citation by The Overseas Press Club. He is a graduate of Kenyon College, The University of London, and was a Knight-Bagehot Fellow at Columbia University.


John Markoff

Topic 6: AI to improve patients’ ability to self-assess symptoms, select and implement self-care options to avoid use of or more successfully partner with health care professionals

John Markoff is currently researching a biography of Stewart Brand, the creator of the Whole Earth Catalog. He is a research affiliate at the Center for Advanced Study in the Behavioral Sciences in 2018-2019, participating in projects focusing on the future of work and artificial intelligence.


Previously he was a reporter at the New York Times, beginning in March 1988 as the paper’s national computer writer. He moved to Silicon Valley to write about technology in 1992. Prior to joining the Times, he worked for The San Francisco Examiner from 1985 to 1988. He reported for the New York Times Science Section from 2010 through 2015. He returned to the Business Section to cover Silicon Valley in 2016 and retired from the paper in December of 2016.

He has also been a lecturer at the University of California at Berkeley School of Journalism and an adjunct faculty member of the Stanford Graduate Program on Journalism.

In 2005, with a group of Times reporters, he received the Loeb Award for business journalism. In 2007 he shared the Society of American Business Editors and Writers Breaking News award. In 2013 he was awarded a Pulitzer Prize in explanatory reporting as part of a New York Times project on labor and automation.


Paul Saffo

Topic 5: Innovation in AI methods that increase AI’s capacity to improve healthcare

Paul Saffo is a Silicon Valley-based forecaster with three decades experience helping corporate and governmental clients understand and respond to the dynamics of large-scale, long-term change.

He teaches forecasting in the Engineering School at Stanford, and is chair of Future Studies at Singularity University. Paul is also a non-resident Senior Fellow at the Atlantic Council, and a Fellow of the Royal Swedish Academy of Engineering Sciences. Paul holds degrees from Harvard College, Cambridge University, and Stanford University.





Hosted By

In Collaboration With

Questions? Email sskelly@stanford.edu for more information.