2020 Research Papers

POSTERS A

1. Learning to Ask Medical Questions using Reinforcement Learning
   Uri Shaham (Yale University); Tom Zahavy (DeepMind); Daisy Massey (Yale University); Shiwani Mahajan (Yale University); Cesar Caraballo (Yale University); Harlan Krumholz (Yale University)

2. ScanMap: Supervised Confounding Aware Non-negative Matrix Factorization for Polygenic Risk Modeling
   Yuan Luo (Northwestern University); Chengsheng Mao (Northwestern University)

3. An Evaluation of the Doctor-Interpretability of Generalized Additive Models with Interactions
   Stefan Hegselmann (University of Münster); Thomas Volkert (University Hospital Münster); Hendrik Ohlenburg (University Hospital Münster); Antje Gottschalk (University Hospital Münster); Martin Dugas (University of Münster); Christian Ertmer (University Hospital Münster)

4. Towards Early Diagnosis of Epilepsy from EEG Data
   Diyuan Lu (Frankfurt Institute for Advanced Studies); Sebastian Bauer (Neurology and Epilepsy Center Frankfurt Rhine-Main, University Hospital Goethe-University); Valentin Neubert (Universitätsmedizin Rostock, Oscar-Langendorff-Institut für Physiologie, Rostock); Lara Costard (Tissue Engineering Research Group, Royal College of Surgeons Ireland); Felix Rosenow (Neurology and Epilepsy Center Frankfurt Rhine-Main, University Hospital Goethe-University); Jochen Triesch (Frankfurt Institute for Advanced Studies)

5. Developing Personalized Models of Blood Pressure Estimation from Wearable Sensors Data Using Minimally-trained Domain Adversarial Neural Networks
   Lida Zhang (Texas A&M University); Nathan Hurley (Texas A&M University); Bassem Ibrahim (Texas A&M University); Erica Spatz (Yale University); Harlan Krumholz ( Center for Outcomes Research and Evaluation / Yale University); Roozbeh Jafari (Texas A&M University); Bobak J Mortazavi (Texas A&M University)

6. Optimizing Influenza Vaccine Composition: A Machine Learning Approach
   Hari Bandi (MIT); Dimitris Bertsimas (MIT)

7. Towards data-driven stroke rehabilitation via wearable sensors and deep learning
   Aakash Kaku (NYU Center for Data Science); Avinash Parnandi (NYU School of Medicine); Anita Venkatesan (NYU School of Medicine); Natasha Pandit (NYU School of Medicine); Heidi Schambra (NYU School of Medicine); Carlos Fernandez-Granda (NYU)

8. Learning Insulin-Glucose Dynamics in the Wild
   Andy Miller (Apple); Nicholas Foti (Apple); Emily Fox (Apple)

9. Knowledge-Base Completion for Constructing Problem-Oriented Medical Records
   James Mullenbach (ASAPP); Jordan Swartz; Greg McKelvey (ASAPP); Hui Dai (ASAPP); David Sontag (ASAPP)

10. Neural Conditional Event Time Models
    Matthew Engelhard (Duke University); Samuel Berchuck (Duke University); Joshua D'Arcy (Duke University); Ricardo Henao (Duke University)

11. Dynamically Extracting Outcome-Specific Problem Lists from Clinical Notes with Guided Multi-Headed Attention
    Justin Lovelace (Texas A&M University); Nathan Hurley (Texas A&M University); Adrian Haimovich (Yale University); Bobak J Mortazavi (Texas A&M University)

12. Differentially Private Survival Function Estimation
    Lovedeep Singh Gondara (Simon Fraser University); Ke Wang (Simon Fraser University)

13. Rotator Cuff Tears Diagnosis Using Weighted Linear Combination and Deep Learning
    Mijung Kim (Ghent University); Ho-min Park (Ghent University); Jae Yoon Kim (Chung-Ang University Hospital); Seong Hwan Kim (Chung-Ang University Hospital); Sofie Van Hoeke (Ghent University); Wesley De Neve (Ghent University)

14. Personalized input-output hidden Markov models for disease progression modeling
    Kristen Severson (IBM Research); Lana Chahine (University of Pittsburgh); Luba Smolensky (Michael J. Fox Foundation); Kenney Ng (IBM Research); Jianying Hu (IBM); Soumya Ghosh (IBM Research)

15. Phenotyping with Prior Knowledge
    Asif Rahman (Philips Research North America); Yale Chang (Philips Research North America); Bryan Conroy (Philips Research North America); Minnan Xu-Wilson ( Philips Research North America)

16. Addressing Sample Size Challenges in Linked Data Through Data Fusion
    Srikesh Arunajadai (Kantar Inc.); Lulu Lee (Kantar Inc.); Tom Haskell (Kantar Inc.)

17. A Causally Formulated Hazard Ratio Estimation through Backdoor Adjustment on Structural Causal Model
    Riddhiman Adib (Marquette University); Paul Griffin (Regenstrief Center for Healthcare Engineering); Sheikh Ahamed (Marquette University); Mohammad Adibuzzaman (Regenstrief Center for Healthcare Engineering)

18. Comparisons Between Hamiltonian Monte Carlo and Maximum A Posteriori For A Bayesian Model For Apixaban Induction Dose & Dose Personalization
    Demetri Pananos (Western University); Daniel Lizotte (UWO)

19. Evaluating and interpreting caption prediction for histopathology images
    Renyu Zhang (University of Chicago); Robert Grossman (University of Chicago); Christopher Weber (University of Chicago); Aly Khan ( Toyota Technological Institute at Chicago);

20. Students Need More Attention: BERT-based Attention Model for Small Data with Application to Automatic Patient Message Triage
    Shijing Si (Duke University); Rui Wang (Duke University); Jedrek Wosik (Duke SOM); Hao Zhang (Duke University); David Dov (Duke University); Guoyin Wang (Duke University); Ricardo Henao (Duke University); Lawrence Carin Duke (CS)

21. Attentive Adversarial Network for Large-Scale Sleep Staging
    Samaneh Nasiri Ghosheh Bolagh (Emory University); Gari Clifford (Department of Biomedical Engineering, Emory School of Medicine)

22. Using deep networks for scientific discovery in physiological signals
    Uri Shalit (Technion); Danny Eytan (Technion); Bar Eini Porat (Technion, Israel institute of technology); Tom Beer (Technion)

POSTERS B

1. Attention-based network for weak labels in neonatal seizure detection
   Dmitry Yu Isaev (Duke University); Dmitry Tchapyjnikov (Duke University); MIchael Cotten (Duke University); David Tanaka (Duke University); Natalia L Martinez (Duke University); Martin A Bertran (Duke University); Guillermo Sapiro (Duke University); David Carlson (Duke University)

2. Deep Reinforcement Learning for Closed-Loop Blood Glucose Control
   Ian Fox (University of Michigan); Joyce Lee (University of Michigan); Rodica Busui (University of Michigan); Jenna Wiens (University of Michigan)

3. Deep Kernel Survival Analysis and Subject-Specific Survival Time Prediction Intervals
   George H Chen (Carnegie Mellon University)

4. Time-Aware Transformer-based Network for Clinical Notes Series Prediction
   Dongyu Zhang (Worcester Polytechnic Institute); Jidapa Thadajarassiri (Worcester Polytechnic Institute); Cansu Sen (WPI); Elke Rundensteiner (WPI)

5. Transfer Learning from Well-Curated to Less-Resourced Populations with HIV
   Sonali Parbhoo (Harvard University); Mario Wieser (University of Basel); Volker Roth (University of Basel); Finale Doshi-Velez (Harvard)

6. Towards an Automated SOAP Note: Classifying Utterances from Medical Conversations
   Benjamin J Schloss (Abridge AI); Sandeep Konam (Abridge AI)

7. Query-Focused EHR Summarization to Aid Imaging Diagnosis
   Denis J McInerney (Northeastern); Borna Dabiri (Brigham and Women's Hospital); Anne-Sophie Touret (Brigham and Women's Hospital); Geoffrey Young (Brigham and Women's Hospital, Harvard Medical School); Jan-Willem van de Meent (Northeastern University); Byron Wallace (Northeastern)

8. Predicting Drug Sensitivity of Cancer Cell Lines via Collaborative Filtering with Contextual Attention
   Yifeng Tao (Carnegie Mellon University); Shuangxia Ren (University of Pittsburgh); Michael Ding (University of Pittsburgh); Russell Schwartz (Carnegie Mellon University); Xinghua Lu (University of Pittsburgh)

9. Hidden Risks of Machine Learning Applied to Healthcare: Unintended Feedback Loops Between Models and Future Data Causing Model Degradation
   George A Adam (University of Toronto); Chun-Hao Chang (University of Toronto); Benjamin Haibe-Kains (University Health Network); Anna Goldenberg (University of Toronto)

10. Self-Supervised Pretraining with DICOM metadata in Ultrasound Imaging
    Szu-Yeu Hu (Massachusetts General Hospital); Shuhang Wang (Massachusetts General Hospital); Wei-Hung Weng (MIT); Jingchao Wang (Massachusetts General Hospital); Xiaohong Wang (Massachusetts General Hospital); Arinc Ozturk (Massachusetts General Hospital); Qian Li (Massachusetts General Hospital); Viksit Kumar (Massachusetts General Hospital); Anthony Samir (MGH/MIT Center for Ultrasound Research & Translation)

11. Deep Learning Applied to Chest X-Rays: Exploiting and Preventing Shortcuts
    Sarah Jabbour (University of Michigan); David Fouhey (University of Michigan); Ella Kazerooni (University of Michigan ); Michael Sjoding (University of Michigan); Jenna Wiens (University of Michigan)

12. Clinical Collabsheets: 53 Questions to Guide a Clinical Collaboration
    Shems Saleh (Vector Institute); Willie Boag (MIT); Lauren Erdman (SickKids Hospital, Vector Institute, University of Toronto); Tristan Naumann (Microsoft Research Redmond, US)

13. Non-invasive Classification of Alzheimer's Disease Using Eye Tracking and Language
    Hyeju Jang (University of British Columbia); Oswald Barral (The University of British Columbia); Giuseppe Carenini (University of British Columbia); Cristina Conati (University of British Columbia); Thalia Field (University of British Columbia); Thomas Soroski (University of British Columbia); Sheetal Shajan (University of British Columbia); Sally Newton-Mason (University of British Columbia)

14. Fast, Structured Clinical Documentation via Contextual Autocomplete
    Divya Gopinath (MIT); Monica N Agrawal (MIT); Luke Murray (MIT); Steven Horng (BIDMC); David Karger (MIT); David Sontag (MIT)

15. Comparing Machine Learning Techniques for Blood Glucose Forecasting Using Free-living and Patient Generated Data
    Hadia Hameed (Stevens Institute of Technology); Samantha Kleinberg (Stevens Institute of Technology)

16. UPSTAGE: Unsupervised Context Augmentation for Utterance Classification in Patient-Provider Communication
    Do June Min (University of Michigan); Veronica Perez-Rosas (UMich); Stanley Kuo (University of Michigan); William Herman (University of Michigan); Rada Mihalcea (University of Michigan)

17. ChexBERT: Approximating the CheXpert labeler for Speed, Differentiability, and Probabilistic Output
    Matthew BA McDermott (MIT); Tzu-Ming H Hsu (MIT); Wei-Hung Weng (MIT); Marzyeh Ghassemi (University of Toronto, Vector Institute); Peter Szolovits (MIT)

18. Robust Benchmarking for Machine Learning of Clinical Entity Extraction
    Monica N Agrawal (MIT); Chloe O'Connell (Partners HealthCare); Ariel Levy (MIT); Yasmin Fatemi (Partners HealthCare); David Sontag (MIT)

19. Preparing a Clinical Support Model for Silent Mode in General Internal Medicine
    Bret Nestor* (University of Toronto); Liam G. McCoy* (University of Toronto); Amol Verma (SMH); Chloe Pou-Prom (SMH); Joshua Murray (SMH), Sebnem Kuzulugil (SMH), David Dai (SMH), Muhammad Mamdani (SMH), Anna Goldenberg (University of Toronto, Vector Institute, SickKids); Marzyeh Ghassemi (University of Toronto, Vector Institute)

    *denotes equal contribution

2020 Clinical Abstracts

Conflict of Interest Statement - Public trust in the peer review process and the credibility of published articles depend in part on how well conflict of interest is handled during writing, peer review, and editorial decision making. Conflict of interest exists when an author (or the author's institution), reviewer, or editor has financial or personal relationships that inappropriately influence (bias) his or her actions (such relationships are also known as dual commitments, competing interests, or competing loyalties). These relationships vary from those with negligible potential to those with great potential to influence judgment, and not all relationships represent true conflict of interest. The potential for conflict of interest can exist whether or not an individual believes that the relationship affects his or her scientific judgment. Financial relationships (such as employment, consultancies, stock ownership, honoraria, paid expert testimony) are the most easily identifiable conflicts of interest and the most likely to undermine the credibility of the journal, the authors, and of science itself. However, conflicts can occur for other reasons, such as personal relationships, academic competition, and intellectual passion.

Statement of Informed Consent - Patients have a right to privacy that should not be infringed without informed consent. Identifying information, including patients' names, initials, or hospital numbers, should not be published in written descriptions, photographs, and pedigrees unless the information is essential for scientific purposes and the patient (or parent or guardian) gives written informed consent for publication. Informed consent for this purpose requires that a patient who is identifiable be shown the manuscript to be published. Authors should identify Individuals who provide writing assistance and disclose the funding source for this assistance.

Identifying details should be omitted if they are not essential. Complete anonymity is difficult to achieve, however, and informed consent should be obtained if there is any doubt. For example, masking the eye region in photographs of patients is inadequate protection of anonymity. If identifying characteristics are altered to protect anonymity, such as in genetic pedigrees, authors should provide assurance that alterations do not distort scientific meaning and editors should so note.

The requirement for informed consent should be included in the journal's instructions for authors. When informed consent has been obtained it should be indicated in the published article.

Statement of Human and Animal Rights - When reporting experiments on human subjects, authors should indicate whether the procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000 (5). If doubt exists whether the research was conducted in accordance with the Helsinki Declaration, the authors must explain the rationale for their approach, and demonstrate that the institutional review body explicitly approved the doubtful aspects of the study. When reporting experiments on animals, authors should be asked to indicate whether the institutional and national guide for the care and use of laboratory animals was followed.