Research Papers

  1. Few-Shot Learning for Dermatological Disease Diagnosis
    Viraj Prabhu (Georgia Tech); Anitha Kannan (Curai); Murali Ravuri (Curai); Manish Chablani (Curai); David Sontag (MIT); Xavier Amatriain (Curai)

  2. Thyroid Cancer Malignancy Prediction From Whole Slide Cytopathology Images
    David Dov (Duke University); Shahar Kovalsky (Duke University); Jonathan Cohen (Duke University); Danielle Range (Duke University); Ricardo Henao (Duke University); Lawrence Carin Duke (Duke University)

  3. ASAC: Active Sensing using Actor-Critic models
    Jinsung Yoon (UCLA); James Jordon (University of Oxford); Mihaela van der Schaar (UCLA and University of Cambridge)

  4. Multimodal Machine Learning for Automated ICD Coding
    KEYANG XU (Petuum Inc.); Mike Lam (Petuum Inc.); Jingzhi Pang (Petuum Inc.); Xin Gao (Petuum Inc.); Charlotte Band (Petuum Inc.); Piyush Mathur (Cleveland Clinic); Frank Papay (Cleveland Clinic); Ashish Khanna (Cleveland Clinic); Jacek Cywinski (Cleveland Clinic); Kamal Maheshwari ( Cleveland Clinic); Pengtao Xie (Petuum / CMU); Eric Xing (Petuum Inc.)

  5. Clinical Question Answering from Electronic Health Records
    Bhanu Pratap Singh Rawat (UMass Amherst); Fei Li (UMASS Lowell); Hong Yu (University of Massachusetts)

  6. Multi-view Multi-task Learning for Improving Autonomous Mammogram Diagnosis
    Trent M Kyono (UCLA); Fiona Gilbert (University of Cambridge); Mihaela van der Schaar (UCLA and University of Cambridge)

  7. Self-Attention Based Molecule Representation for Predicting Drug-Target Interaction Bonggun Shin (Emory University); Sungsoo Park (Deargen Inc.); Keunsoo Kang (Dankook University); Joyce Ho (Emory University)

  8. Enhancing high-content imaging for studying microtubule networks at large-scale
    Hao-Chih Lee (MSSM)

  9. Measuring the Sympathetic Response to Intense Exercise in a Practical Setting
    Shiva Kaul (CMU)

  10. Early Recognition of Sepsis with Gaussian Process Temporal Convolutional Networks and Dynamic Time Warping
    Michael Moor (MLCB, D-BSSE, ETH Zurich); Max Horn (MLCB, D-BSSE, ETH Zurich); Bastian A Rieck (MLCB, D-BSSE, ETH Zurich); Damian Roqueiro (ETH Zurich); Karsten Borgwardt (ETH)

  11. Simultaneous Measurement Imputation and Outcome Prediction for Achilles Tendon Rupture Rehabilitation
    Charles Hamesse (Royal Military Academy, Belgium); Ruibo Tu (KTH Royal Institute of Technology)*; Paul Ackermann (Karolinska Institutet); Hedvig Kjellström (KTH Royal Institute of Technology); Cheng Zhang (Microsoft)

  12. What Clinicians Want: Contextualizing Explainable Machine Learning for Clinical End Use Sana Tonekaboni (University of Toronto, Vector institute, Hospital for Sick children); Shalmali Joshi (Vector Institute); Melissa Mccradden (Hospital for Sick children, Vector institute); Anna Goldenberg (University of Toronto, Hospital for Sick Children)

  13. “Brilliant” or “Beautiful”? Patient Perceptions of Male versus Female Physicians in Online Reviews
    Avijit Thawani (); Michael Paul (CU Boulder); Urmimala Sarkar (UC San Francisco); Byron Wallace (Northeastern University)

  14. Relaxed Weight Sharing: Effectively Modeling Time-Varying Relationships in Clinical Time-Series
    Jeeheh Oh (University of Michigan); Jiaxuan Wang (University of Michigan); Shengpu Tang (University of Michigan); Michael Sjoding (University of Michigan); Jenna Wiens (University of Michigan)
    *equal contribution

  15. Clinically Accurate Chest X-Ray Report Generation
    Tzu-Ming H Hsu (MIT); Guanxiong Liu (University of Toronto); Matthew BA McDermott (MIT); Willie Boag (MIT); Wei-Hung Weng (MIT); Pete Szolovits (MIT); Marzyeh Ghassemi (University of Toronto)

  16. Automated Estimation of Food Type from Body-worn Audio and Motion Sensors in Free-Living Environments
    Mark Mirtchouk (Stevens Institute of Technology); Dana McGuire (Stevens Institute of Technology); Andrea Deierlein (New York University); Samantha Kleinberg (Stevens Institute of Technology)

  17. FLARe: Forecasting by Learning Anticipated Representations
    Yeahuay Wu (UMass Amherst); Surya Teja Devarakonda (UMass Amherst); Madalina Fiterau (University of Massachusetts Amherst)

  18. Multi-Task Gaussian Processes and Dilated Convolutional Networks for Reconstruction of Reproductive Hormonal Dynamics
    Tristan Bertin (Columbia University); Theresa Hardy (New York University); David Albers (University of Colorado); Iñigo Urteaga (Columbia University); Noémie Elhadad (Columbia)*

  19. A Neural Model for Predicting Dementia from Language
    Weirui Kong (University of British Columbia); Hyeju Jang (University of British Columbia); Giuseppe Carenini (University of British Columbia); Thalia Field (University of British Columbia)

  20. A calibration metric for risk scores with survival data
    Steve Yadlowsky (Stanford University); Sanjay Basu (Stanford University); Lu Tian (Stanford University)

  21. Multiple Instance Learning for ECG Risk Stratification
    Divya Shanmugam (MIT); John Guttag (MIT); Davis Blalock (MIT)

  22. Using Contextual Information to Improve Blood Glucose Prediction
    Mohammad Akbari (University College London); Rumi Chunara (New York University)

  23. Embryo Staging with Weakly-Supervised Region Selection and Dynamically-Decoded Predictions
    Tingfung Lau (Carnegie Mellon University); Nathan Ng (UCSD); Julian A Gingold (Cleveland Clinic Foundation); Nina Desai (Cleveland Clinic Foundation); Julian McAuley (UCSD); Zachary Lipton (Carnegie Mellon University)

  24. Using Domain Knowledge to Overcome Latent Variables in Causal Inference from Time Series
    Min Zheng (Stevens Institute of Technology); Samantha Kleinberg (Stevens Institute of Technology)

  25. Dynamically Personalized Detection of Hemorrhage
    Chirag Nagpal (Carnegie Mellon University); Xinyu Li (Carnegie Mellon University); Artur Dubrawski (CMU)

  26. Temporal Graph Convolutional Networks for Automatic Seizure Detection
    Ian Covert (University of Washington); Jiening Zhan (Google); Matthew Shore (Google); Ming Jack Po (Google)

  27. A Spatiotemporal Approach to Predicting Glaucoma Progression Using a CT-HMM
    Supriya Nagesh (Georgia Institute of Technology); Alexander F Moreno (Georgia Institute of Technolog); James Rehg (Georgia Institute of Technology)

  28. Counterfactual Reasoning for Fair Clinical Risk Prediction
    Stephen R Pfohl (Stanford University); Tony Duan (Stanford University); Daisy Yi Ding (Stanford University); Nigam Shah (Stanford)

  29. Learning from Few Subjects with Large Amounts of Voice Monitoring Data
    Jose Javier Gonzalez Ortiz (MIT)*; Daryush Mehta (MGH); Jarrad Van Stan (MGH); Robert Hillman (MGH); John Guttag (MIT); Marzyeh Ghassemi (University of Toronto)

  30. Meta-Weighted Gaussian Process Experts for Personalized Forecasting of AD Cognitive Changes
    Ognjen Rudovic (MIT); Yuria Utsumi (MIT); Kelly Peterson (MIT); Ricardo Guerrero (Imperial College ); Daniel Rueckert (Imperial College London); Rosalind Picard (MIT)

  31. SLEEPER: interpretable Sleep staging via Prototypes from Expert Rules
    Irfan Al-Hussaini (Georgia Institute of Technology); Cao Xiao (IQVIA); Brandon Westover (Mass General); Jimeng Sun (Georgia Tech)

  32. EEG2Text: Learning to Write Medical Reports from EEG Recordings
    Siddharth Biswal (Georgia Institute of Technology); Cao Xiao (IQVIA); Brandon Westover (Mass General); Jimeng Sun (Georgia Tech)

  33. Predicting Sick Patient Volume in a Pediatric Outpatient Setting using Time Series Analysis Grace Guan (Princeton University); Barbara Engelhardt (Princeton University)

  34. Feature Robustness in Non-stationary Health Records: Caveats to Deployable Model Performance in Common Clinical Machine Learning Tasks
    Bret Nestor (University of Toronto); Matthew BA McDermott (MIT); Willie Boag (MIT); Gabriela Berner (); Tristan Naumann (Microsoft Research); Michael C Hughes (Tufts University); Anna Goldenberg (University of Toronto, Hospital for Sick Children); Marzyeh Ghassemi (University of Toronto)

  35. The Medical Deconfounder: Assessing Treatment Effect with Electronic Health Records (EHRs)
    Linying Zhang (Columbia University); Yixin Wang (Columbia University); Anna Ostropolets (Columbia University); Jami Mulgrave (Columbia University); David Blei (Columbia University); George Hripcsak (Columbia University)

  36. Predicting Phase 3 Clinical Trial Results by Modeling Phase 2 Clinical Trial Subject Level Data Using Deep Learning
    Youran Qi (University of Wisconsin–Madison); Qi Tang (Sanofi US)

  37. Phenotype Inference with Semi-Supervised Mixed Membership Models
    Victor A Rodriguez (Columbia University); Adler Perotte (Columbia University)


Clinical Abstracts

  1. Examining the measurement of quality in healthcare using artificial intelligence methods: a study of quality in long-term care Andrea Iaboni (Toronto Rehabilitation Institute)*; Pouria Mashouri (University Health Network); Babak Taati (University Health Network)

  2. Prediction of the next medication order to assist prescription verification by pharmacists in a health care center Maxime Thibault (CHU Sainte-Justine)*

  3. Correlating ECG features with symptom burden in patients with atrial fibrillation using Markov modeling Hamid Ghanbari (University of Michigan Department of Cardiology); Kevin M Wheelock (University of Michigan Medical School)*

  4. An Evaluation Toolkit to Guide Model Selection and Cohort Definition in Causal Inference Yishai Shimoni (IBM Research - Haifa)*; Sharon Hensley Alford (IBM Watson Health); Yaara Goldschmidt (IBM)

  5. Developing Production-Ready, Artificially Intelligent mHealth Tools: Sentiment Analysis to Track Positivity Anne S Morrow (Nova Southeastern University)*; Alexandro Campos Vega (Zyanya Tech, LLC); Xin Zhao (Florida International University); Katherine McCurry (Virginia Tech); Michelle Liriano (Florida International University)

  6. Stimulated Raman histology and deep neural networks for near real-time intraoperative brain tumor diagnosis Todd C Hollon (1985)*; Balaji Pandian (U); Christian Freudiger (Invenio Imaging ); Petet Canoll (Columbia University); Honglak Lee (UM); Daniel Orringer (Univer)

  7. Deep Learning Classification of Home Sleep Apnea Test Michael B Wilson (University of Michigan)*

  8. Modeling heparin protocol dosing compliance using dynamic and static data to improve clinical outcomes Aaron Noll (Cerner)*; J. Marc Overhage (Cerner Corporation); Bennett Lovejoy (Cerner Corporation); Jeremy Foster (Cerner Corporation)

  9. A computational approach to quantify movement impairment after stroke Avinash Parnandi (NYU School of Medicine)*; Adisa Velovic (NYU); Audre Wirtanen (Balance Arts Center); Heidi Schambra (NYU School of Medicine)

  10. Refining movement quantitation in stroke Avinash Parnandi (NYU School of Medicine)*; Jasim Uddin (CUMC); Dawn Nilsen (CUMC); Heidi Schambra (NYU School of Medicine)

  11. Novel Leveraging of Perioperative Data for Early Diagnosis of Heart Failure: A Machine Learning Approach Michael Mathis (University of Michigan)*; Hyeon Joo (University of Michigan); Milo Engoren (University of Michigan); Brahmajee Nallamothu (University of Michigan); Michael Burns (University of Michigan - MPOG); Kayvan Najarian (University of Michigan); Sachin Kheterpal (University of Michigan)

  12. Privacy Pitfalls in Tree-Based Ensemble Models Developed using Electronic Health Record Data: A Case Study Karandeep Singh (University of Michigan)*; Adharsh Murali (University of Michigan); Michael Burns (University of Michigan - MPOG); Brahmajee Nallamothu (University of Michigan); Akbar Waljee (University of Michigan); Jeremy Sussman (University of MIchigan)

  13. The 7 Habits of Effective Predictive Model Implementations: Lessons from the Clinical Trenches Karandeep Singh (University of Michigan)*; Sean Meyer (University of Michigan); Michael Burns (University of Michigan - MPOG); Jeremy Sussman (University of MIchigan); Akbar Waljee (University of Michigan); Brahmajee Nallamothu (University of Michigan)

  14. Predicting pediatric extubation failure with machine learning methods through an innovative tandem approach Sydney Rooney (Pittsburgh University); Evan L Reynolds (University of Michigan)*; Michael Gaies (University of Michigan); Mousumi Banerjee (University of Michigan)

  15. Democratizing EHR Analyses - A Comprehensive, Generalizable Pipeline for Learning from Clinical Data Michael Sjoding (University of Michigan)*; Shengpu Tang (University of Michigan); Parmida Davarmanesh (University of Michigan); Yanmeng Song (University of Michigan); Danai Koutra (U Michigan); Jenna Wiens (University of Michigan)

  16. Leveraging Machine Learning to Decrease In-Hospital Mortality Rates Nathan Brajer (Duke University School of Medicine)*

  17. Return to Work After Injury: A Sequential Prediction & Prescription Problem Erkin Otles (University of Michigan)*; Haozhu Wang (‎University of Michigan); Suynapeng Zhang (University of Michigan); Brian Denton (University of Michigan); Jon Seymour (Peers Health); Jenna Wiens (University of Michigan)

  18. Predicting ICU length of stay with a competing risk deep learning survival model framework Xinggang Liu (Philips); Omar Badawi (Philips)*

  19. Using Predictive Mortality and Cardiogenic Shock Identification Tools to Support Team-Based Treatment Intervention on Adult Cardiology Patients at Duke University Hospital William Ratliff (Duke Institute for Health Innovation)*; Aman Kansal (Duke Institute for Health Innovation); Sehj Kashyap (Duke Institute for Health Innovation)

  20. Model Ensembling vs Data Pooling: Alternative ways to merge hospital information across sites Kristin M Corey (Duke Institute for Healthcare Innovation)*; Elizabeth Lorenzi (Duke University Department of Statistical Science); Michael Gao (Duke Institute for Health Innovation); Suresh Balu (Duke Institute for Healthcare Innovation); Mark Sendak (Duke University)

  21. Improving Readmissions Modeling with Social Determinants of Health Joshua Helmkamp (Duke Institute for Health Innovation); Heather Rosett (Duke Institute for Health Innovation); Morgan G Simons (Duke University School of Medicine); Mark Sendak (Duke University); Suresh Balu (Duke Institute for Healthcare Innovation); Allan Kirk (allan.kirk@duke.edu); Kristin M Corey (Duke Institute for Healthcare Innovation)*

  22. New Methods of Natural Language Processing using Machine Learning Methods to Identify Ischemic Stroke Presence, Acuity and Location from Clinical Radiology Reports Charlene J Ong (Boston University)*

  23. Complex patient phenotypes in critically care associate with high mortality rates in sepsis. Zsolt E Zador (University of Toronto)*; Alexander Landry (University of Toronto); Julian Spears (University of Toronto); Nophar Geifman (University of Toronto)

  24. Reconsidering Missingness: Seeking Implicit Clinical Insights in Sparse Medical Variables Liam G McCoy (University of Toronto)*; Marzyeh Ghassemi ()

  25. Errors in Vital Sign Recording Within a Publicly Accessible ICU Research Database Heather O'Halloran (Queens University)*; David Maslove (Queens University); Richard Veldhoen (Queens University)

  26. Machine Learning Medical Directives at Triage in Pediatric Emergency Medicine: The First Step to Automated Pathways for Healthcare Delivery Devin Singh (SickKids), Carson McLean (University of Toronto), Lauren Erdman (University of Toronto), Lebo Radebe (University of Toronto), Erik Drysdale (University of Toronto), Jason Fischer (SickKids), Anna Goldenberg (SickKids), and Michael Brudno (SickKids)

  27. Emergency Department Sepsis Alert System for Children Devin Singh (SickKids), Carson McLean (University of Toronto), Lebo Radebe (SickKids), Lauren Erdman (University of Toronto), Erik Drysdale (University of Toronto), Michael Brudno (SickKids), Anna Goldenberg (SickKids)

  28. Anti-Diabetic Drug Repurposing Using Electronic Health Records: Design, Emulation and Analysis of a Synthetic In-Silico Clinical Trial for Alzheimer’s Disease Stan Finkelstein (MIT)*; Shenbo Xu (MIT); Bowen Su (Imperial College London); Bang Zheng (Imperial College London); Marie-Laure Charpignon (MIT); Ioanna Tzoulaki (Imperial College London); Lefkos Middleton (Imperial College London); Roy Welsch (MIT)

  29. Development of a Clinical Decision Tool and Protocol for Identification and Treatment of Corticosteroid Induced Hyperglycemia Morgan G Simons (Duke University School of Medicine)*; Joseph Futoma (Duke); Kristin Corey (Duke SOM); Michael Gao (Duke Institute for Health Innovation); Marshall Nichols (Duke Institute for Health Innovation); Krista Whalen (Duke Institute for Healthcare Innovation); Mark Sendak (Duke University); Finale Doshi-Velez (Harvard); Ann McGee (Duke University School of Medicine); Tracu Setji (Duke University School of Medicine)