AI Health Postdoctoral Associate, Duke University

The International Society for Bayesian Analysis; Durham (NC); On-site; USD 60,000 - 75,000; Full time

Job summary

Duke University's Berchuck lab seeks a motivated Postdoctoral Associate in applied Bayesian modeling and machine learning. This two-year position involves analyzing large longitudinal datasets, contributing to projects in glaucoma and cancer research, with opportunities for publication and interdisciplinary collaboration.

Qualifications

• Proficiency in Bayesian inference techniques (e.g., MCMC, Stan).
• Experience with electronic health records and advanced statistical computing.
• Track record of first-author publication in peer-reviewed journals.

Responsibilities

• Analyze longitudinal biomedical data using Bayesian modeling.
• Develop algorithms for predicting psychosocial distress and identifying pain trajectories.
• Engage in applied methods development for high-dimensional longitudinal settings.

Skills

Bayesian inference techniques

Longitudinal data analysis

Machine learning approaches

Statistical computing

Communication skills

Education

Doctorate in biostatistics, statistics, machine learning, or related discipline

Job description

AI Health Postdoctoral Associate, Duke University

Oct 12, 2023

The Berchuck lab at Duke University (Durham, NC, USA) is looking for a highly motivated Postdoctoral Associate in the fields of applied Bayesian modeling and machine learning using large longitudinal biomedical data, including electronic health records and mobile health data. The position will be funded by Samuel I. Berchuck, PhD who holds appointments in the Departments of Biostatistics & Bioinformatics and Statistical Science and is a faculty affiliate of Duke AI Health. The appointment is for 2 years, starting October 1, 2023, or a later date, and may be renewed for 1 year. The postdoc will have the opportunity to be affiliated with Duke AI Health as a Duke AI Health Postdoc.

OVERVIEW. The successful applicant will use tools from statistics, biostatistics, and machine learning to analyze and model rich longitudinal datasets from patients with chronic diseases, including glaucoma and cancer. Possible projects include (i) developing Bayesian semi-supervised inference algorithms for predicting psychosocial distress in patients with glaucoma, (ii) identifying pain trajectories in cancer survivors using Bayesian latent class models, and (iii) developing scalable Bayesian inference algorithms for high-dimensional longitudinal settings. All three of these projects will require a certain degree of applied methods development.

QUALIFICATIONS. The successful applicant has a doctorate in biostatistics, statistics, machine learning, computer science, or a related discipline. The position requires proficiency in Bayesian inference techniques (e.g., MCMC, Stan), longitudinal data analysis, and modern machine learning approaches. Extensive experience with real-world (e.g., electronic health records) data and advanced statistical computing skills are required. The successful applicant demonstrates great attention to detail, works independently, takes initiative, has excellent written and verbal communication skills, and performs well in an interdisciplinary environment at the interface between data science and clinical research. The successful candidate is expected to have a track record of first-author publication in peer-reviewed journals.

APPLICATION. Interested applicants should submit the following documents:
-Cover letter describing research interests and career goals (2 pages or less)
-Curriculum vitae (4 pages or less)
-Copies of 1-2 relevant first-author publications
-A list containing the contact information for 3 references

Completed applications will be reviewed on an ongoing basis until the position is filled, with priority given to applications submitted prior to January 1, 2024.