Research Projects

Suzanne Bakken

Center for Evidence-based Research
The goal of this project is to establish an administrative core and pilot study core to supposrt research related to evidence-based practice in underserved populations.
Mining Complex Clinical Data for Patient Safety Research
The project represents a unique opportunity to apply the most advanced medical language processing system to a large, comprehensive clinical repository to advance patient safety researc. Agency for Healthcare Research and Quality 2001-2004

HARVEST
An intelligent summarization system for the electronic health record.
Phenotypic Characterization of Autism Spectrum Disorders through NLP Analysis
Speech and language analysis of video recordings of patients on the autism disorder spectrum.
HealthTermFinder
Automatic tagging of health terms in news stories.
Structuring and Searching the Content of Health Related Web Posts
Intelligent search tool for breast cancer forums. The search relies on automatically induced psycho-social information conveyed in posts.

Pharmacovigilance using Natural Language Processing, the EHR, and Statistics
This project focuses on research that involves use of natural language processing, statistics, and information theory to acquire knowledge to discover novel adverse drug events. We will also collect and integrate knowledge that is publicly available to har
A Biomedical Natural Language Resource
This project focuses on research in natural language processing involving the EHR and the literature, spanning clinical and biomolecular text.

Discovering and applying knowledge in clinical databases
The long term goal is to develop and apply methods to exploit electronic medical record data to support decisions and generate knowledge. We are currently developing an information theoretic framework for characterizing the electronic health record, using the framework to study the EHR and sampling issues, and using the framework and traditional data mining to answer clinical and informatics questions.
Next-generation electronic health records
Incorporate advanced informatics methods for human-computer interaction, collaboration, knowledge management, communication, and data representation to improve the health care encounter and health promotion.
Systems medicine
Apply modeling and computational techniques to problems in health care such that it integrates knowledge and data across scales to better understand and predict processes, leading to improved health care. Current work is focused on infection control, incorporating team processes and policies, the physiology of infections, and the molecular biology of pathogens.

HITEC Evaluation of Health Information Exchange for Public Health
Evaluation of how grantees in New York State are implementing a public health use case under contract with HEAL NY - Phase 5 Health Information Technology Grants: Advancing Interoperability and Community-wide EHR Adoption
Network Analysis of the Growth of an Emerging Community: Public Health Services and Systems Research
Evaluation of an EHR Implementation: NYC Primary Care Implementation Project
Co-investigator
Evaluation of an EHR Implementation: NYC Primary Care Implementation Project
Co-investigator

Understanding the Evolution of RNA Viruses
For the past several years, there has been a worldwide effort to sequence viral genomes. Evolution is a dynamical process that shape genomes. From the tens of thousands of viruses that have been isolated around the world during the last century, we can start to understand the mutational processes and the selection forces that act on these organisms (computational virology). Dr Rabadan and his colleagues at Columbia are developing tools to analyze genomic data, extracting the relevant information to understand the molecular biology, population genetics, evolution, and epidemiology of these viruses.
New Tools for Pathogen Discovery Through Next Generation Sequencing
We are developing computational tools to identify viral genomic material from High Throughput Sequence Data.
Sequencing Cancer Genomes
In collaboration with researchers at the Institute for Cancer Genetics at Columbia University, we are sequencing the genome of several types of cancers to identify the genetic alterations that could contribute to the development of the disease.
Data mining Electronic Health Records
Electronic Health Records provide a rich source of information about the role of patient history in the development of a disease combined with a framework for epidemiological research.

Searching for access anomalies in clinical audit logs
Discovering patterns in audit logs that capture user access to sensitive data which help identify whether specific access are anomalous, and hence subject to further audit, or not.

Computerized Provider Documentation
NewYork-Presbyterian Hospital is transitioning to electronic documentation of physician/provider notes using the Eclipsys Sunrise Acute Care electronic health record. We are assisting with this effort and evaluating its impact.

Bridging the Semantic Gap between Research Eligiblity Criteria and Clinical Data
The goal is to develop a knowledge base of research eligibility criteria linkable to any clinical data warehouse. The key component is a natural language processing engine for eligibility criteria called EliXR.
CUMC Biobank
The goal of this project is to develop a CUMC biobank using discarded patient blood samples to support personalized medicine.