November 1, 2019

Share Duke News

JOB : AI Health Data Science Fellow


Job Code: 1420, Associate in Research

 

Job Description

 

The Duke AI Health Data Science Fellowship program is a 2-year training program for individuals with a strong quantitative background seeking experience in data science with healthcare applications.  The program is part of the broader AI Health Initiative, a key component of Duke University’s investment in data science as a tool to improve the lives of patients and solve problems that matter in healthcare.

 

The AI Health Data Science Fellowship program is structured to develop individuals early in their careers and provide them with an intense but supportive environment to develop essential skills in healthcare analytics and modern methods in artificial intelligence/machine learning. Fellows will gain hands-on experience with cutting-edge data science methods and healthcare data, while working on interdisciplinary teams to tackle projects that improve the lives of patients. Each team will include a quantitative lead, clinical lead, data engineer, the data science fellow, and students/trainees. The fellow will be mentored and work under the direction of the team’s quantitative lead, an individual with advanced expertise in data science and machine learning, to contribute to meaningful, real-world applications guided by the clinical lead(s). Each team will be embedded within a specific clinical area (e.g., cardiology, pediatrics, or psychiatry) allowing the team members to develop a broad understanding and familiarity with the clinical domain.

 

After their 2-year commitment, fellows can expect to be highly competitive for an array of opportunities in healthcare, academia, or industry.

 

Interested candidates should send an email to plus-datascience@duke.edu with the subject line “AI Health Fellowship Inquiry”. Please include a cover letter and resume.

 

Responsibilities

 

  1. Develop data science and machine learning applications in healthcare (50%)

The Data Science Fellow will collaborate with their team to assess the big picture clinical question, translate the question into an analyzable scope, and frame clear analytic objectives. The fellow will be responsible for all aspects of the analysis, as outlined below.

 

Tasks will include:

  • Developing an analytic plan that accurately reflects the clinical objectives and properly accounts for the population, available data (including the data generating mechanism), data quality and cleanliness, and potential for bias or confounding
  • Working either independently or with a data engineering team to define and create analytic dataset(s)
  • Performing data exploration and descriptive analyses
  • Developing and evaluating appropriate  statistical and machine learning models in a reproducible manner
  • Demonstrating comprehension of statistical and machine learning methods applied, including assumptions and limitations of a given approach

 

  1. Develop tools and generalizable infrastructure (20%)

The Data Science Fellow will ensure the integrity of their results through good documentation and code provenance, carefully managing their code and developing accurate descriptions of what they are doing, and why, in a way that others can understand. Fellows will demonstrate tool-specific skills and apply common software development practices such as modularizing code and using version control (e.g., Git).

 

Tasks will include:

  • Documenting project work with version control and regular project summary reports
  • Developing procedures, modules, and programs to generalize code into reusable resources, within and across projects.
  • Preparing work for transition to implementation and product development teams.

 

  1. Share and disseminate knowledge (20%)

The Data Science Fellow will engage and communicate with their team members, clinical sponsors, and leadership at varying levels of clinical and technical fluency. They will tailor their communications to the appropriate audience, including clinical and quantitative team members, and demonstrate the ability to converse proficiently in many situations with diverse audiences.

 

Tasks will include:

  • Developing content including slides, visualizations, and interactive tools.
  • Writing scientific content such as abstracts and manuscripts.
  • Preparing and delivering presentations.
  • Representing their team at meetings, conferences, and external events.

 

  1. Contribute to the mission of AI Health (10%)

The Data Science Fellow will engage with their teams at a professional level and develop collegial relationships. They will engage as an active contributor with their transdisciplinary team, communicate clearly and regularly, bring issues forward, and work with the team to develop solutions.

 

Tasks will include:

  • With increasing levels of independence, mentoring students and trainees.
  • Participating in the development of new opportunities.

 

Education/Training

Position requires a master’s degree in any quantitative discipline. Examples of quantitative disciplines include (but are not limited to) engineering, computer science, statistics, biostatistics, biomedical engineering, economics.

 

Skills/Experience

  • Demonstrated aptitude for data analysis, programming, and creative problem solving.
  • Demonstrated proficiency in programming, especially Python and R.
  • Experience in statistical modeling is preferable but not required.
  • Experience with version control, databases, distributed computing, or cloud computing is preferable but not required.
  • Intellectual curiosity, as demonstrated by proactive pursuit of learning opportunities and diverse, challenging projects.
  • A passion for the application of data science to solve important problems in healthcare and improving patient lives.