NLP Postdoc at Stanford Center for Population Health Sciences

[Suzanne Tamang]

A two-year postdoctoral position is available at the Stanford Center for Population Health Sciences: http://med.stanford.edu/phs.html Founded in 2015, the Center’s mission is to improve individual and population health by bringing together diverse disciplines and data to understand and address social, environmental, behavioral, and biological factors on both a domestic and global scale.

The postdoctoral scholar will be involved with work that is funded by the US Agency for Healthcare Quality and Research to develop and scale text mining algorithms to extract patient-reported outcomes from clinical notes in the electronic health record. Patient-reported outcomes are key outcomes in rheumatoid arthritis, a condition that affects 1.3 million Americans and is a leading cause of disability in the US. The goal of the project is to develop accurate, valid and scalable text mining algorithms for extracting patient-reported outcomes scores from the Rheumatology Informatics System for Effectiveness (RISE) registry to enhance the feasibility and usability of patient-reported outcome measurement nationally, for rheumatoid arthritis patients.

Since 2014, RISE has been used to facilitate quality improvement on a national scale: https://www.rheumatology.org/I-Am-A/Rheumatologist/Registries/RISE The registry passively extracts electronic health record data from rheumatology practices, aggregates and analyzes these data centrally, and feeds performance on healthcare quality measures back to clinicians using a web-based dashboard. Many health professionals do not have the health-IT support to reconfigure their electronic health record systems to collect patient-reported outcomes as structured data. This results in inability to map this information in RISE, to benchmark patient-reported outcomes data, or to track patient-level data to inform clinical decisions or engage patients in understanding their disease trajectory.

The postdoctoral position also provides the opportunity to work on other important population health problems such as opioid and suicide risk prediction; also, with Danish population health registries.

Prerequisites:
PhD in computer science, statistics, informatics, physics, epidemiology or similar
Excellent programming skills complemented with experience with some combination of statistical modeling, machine learning, natural language processing and informatics
Interest in population health

The ideal candidate will have:
A solid background in a computer science and statistics. Special areas of interest include: natural language processing, data visualization, machine learning, information extraction and application areas such as bioinformatics, population genetics, healthcare or medicine
Excellent problem solving skills
Prior experience working with large corpora
A very good understanding of modern computational methods and algorithms
Expertise in programming languages, such as python, R, C, etc
Team work capacity
Prior experience with health datasets will be a plus
Expertise in deep learning frameworks such as PyTorch or TensorFlow will be a plus

Candidates should send:
A CV, including publications list
A one page motivation letter explaining why their skills, knowledge and experience make them a particularly suitable candidate for the given position
Pointers to their two most important papers or other work

Please send your information to: sta...@stanford.edu

Suzanne Tamang, PhD
Assistant Faculty Director for Data Science, Stanford Center for Population Health Sciences; Instructor, Biomedical Data Science https://med.stanford.edu/phs/phs-people/faculty-leadership.html