Hosted by The Data Science Institute and Columbia University Irving Medical Center.

About the Workshop

DASHI invites applications for pilot projects in the intersection of Artificial Intelligence (AI) and Health Sciences. Towards that end, DASHI will encourage the formation of interdisciplinary project teams through a cross-campus brainstorming workshop.

We encourage health scientists and data scientists across Columbia University to join the workshop and explore pathways for collaboration with selected DASHI project PIs.

Workshop Format: The workshop will include two plenary research presentations from invited speakers; three sessions of lightning talks from selected PIs who are seeking collaborators; and three sessions of breakout groups for PIs to share the scope of their projects and meet with potential collaborators.

Workshop Goal: The goal for this event is to support collaboration, strengthen teams, and advance project summaries before PIs submit a full pilot proposal. Following the workshop, the thus formed team of contact-PI and their selected co-PI(s) may submit a full, 3-page application for seed funding of the pilot project. This proposal should delineate a path towards external funding of the project beyond its pilot stage.

DASHI Budget: DASHI expects to fund at least 2 projects, up to $75k each.

PI Eligibility Criteria: Any faculty member/research scientist across the campus at either Morningside or CUIMC. Postdocs can participate in a collaborative team.

Workshop Date & Time

Friday, December 10, 2021 (9:00 AM – 12:00 PM ET) – Virtual

Registration is open to the Columbia University community. Please email Alexis Avedisian (aa4598@columbia.edu) for an invitation to register.

Agenda

Plenary Session

9:00 AM: Opening Remarks (5 min)

9:05 AM: Invited Talk on Health Sciences: Ali Soroush and Julian Abrams, CUIMC (20 min)

  • Predicting Upper Gastrointestinal Cancer Using EHR Phenotyping

9:25 AM: Invited Talk on Data Science: Nicholas Tatonetti, Data Science Institute (20 min)

  • Talk details coming soon

Session 1: Clinical Scientists Seeking Data Scientists

9:45 AM: Lighting Talks (15 min)

  1. Benjamin Ranard: Improving Sepsis care at NYP Through Improved Measurement and Prediction
  2. Elisa Konofagou: A Novel Machine Learning Approach for the Automated, Ultrasound-based Mapping of Arrhythmia in the Clinic
  3. Fatemeh Momen-Heravi: Development of Informatics Tools for Artificial Intelligence (AI)-Based Pathology Profiling and Integrated Prognostic Modeling
  4. Fereshteh Zandkarimi: Discovery of Lipidomic Signature of Aging in the Brain by Integrating Deep Learning and Mass Spectrometry Imaging
  5. Sandra Albrecht: Use of AI for Spanish-Language Content to Improve Uptake of Mitigation Strategies During Public Heath Emergencies

10:00 AM: BREAKOUT GROUPS. Each lightning talk has their own breakout room (25 min)

5 min buffer

Session 2: Data Scientists Seeking Health Scientists

10:30 AM: Lighting Talks (15 min)

  1. Sharon Di: How Do We Infer Dementia From One’s Driving Trajectories?
  2. Gamze Gursoy: Privacy-Preserving Model Training for Disease Prediction Using Federated Learning With Differential Privacy
  3. Kriste Krstovski: Joint Supervised Topic Embedding Model for EHR Text Data
  4. Linda Valeri: Causal Learning for Digital Psychiatry
  5. Vishal Misra: Accelerating Personalized Drug Discovery using Synthetic Interventions

10:45 AM: BREAKOUT GROUPS. Each lightning talk has their own breakout room (25 min)

5 min buffer

Session 3: Environmental/Behavioral Data Seeking Collaborators

11:15 AM: Lighting Talks (15 min)

  1. Sean Kinney: Leveraging Geological Data for Basic Research in Public Health and Medical Sciences
  2. Anne Grauer: Exploring the Etiology and Demographics of Ordering Errors within the Electronic Health Record
  3. Daniel Westervelt: Linking Personal Air Pollution Exposure to Adverse Health Outcomes in Sub-Saharan Africa
  4. Ipek Ensari: Developing Reinforcement Learning-Based Algorithms to Personalize Exercise Self-Management Regiments in Chronic Pelvic Pain Disorders
  5. Yike Shen: Deep Learning based Network Analysis in the Normative Aging Study

11:30 AM: BREAKOUT GROUPS. Each lightning talk has their own breakout room (25 min)

Closing Remarks

11:55 AM: Proposal guidelines and workshop wrap up (5 min)

Participants will re-group to hear guidelines for submitting a final funding proposal. The link to submit a proposal will be shared in this session. The DASHI committee will answer any questions, recap the workshop, and provide closing statements.

The full pilot proposal deadline is Friday, January 7, 2022.

12:00 PM: Event concludes.

About DASHI

The Data Science Institute and Columbia University Irving Medical Center have a new partnership focused on building collaborative research projects that leverage foundational data science for new clinical advances.On the biomedical side, this is driven by emerging access to large scale complex datasets due to recently deployed technologies, e.g. in imaging, genomics, and electronic health records. These data challenges are looking for analysis tools to tackle them. On the engineering side, method developers are seeking data to sink their teeth into and test their developments in real-world settings. The new Data Science and Health Initiative (DASHI) aims to bridge this gap and create synergy between our institutional strengths.Join the inaugural meeting to meet colleagues and potential collaborators from across Columbia University.

Data Science and Health Initiative Steering Committee Members

  • Itsik Pe’er, PhD, Associate Professor of Computer Science
  • Peter Canoll, MD, PhD, Professor of Pathology & Cell Biology
  • Elham Azizi, PhD, Assistant Professor in Biomedical Engineering
  • Elias Bareinboim, PhD, Associate Professor of Computer Science
  • Lawrence Schwartz, MD, Professor and Chairman for the Department of Radiology
  • Sarah Collins Rossetti, RN, PhD, Assistant Professor of Biomedical Informatics and Nursing