Stanford Clinical Data Science Fellowship Program

Program Summary

A full-time one-year post-doctoral training fellowship (with possibility of a one-year extension) for PhDs and MDs to bridge the gap between data science and clinical medicine. Fellows will receive full salary support from the program. The focus of the program is to immerse fellows in interdisciplinary research and clinical workflows so they can learn, build and deploy real-world health data solutions. The fellowship will be self-directed and project-based, with top researchers and clinician mentors.

Program Details

Fellow Background: Ideal fellows are highly motivated, detail-oriented individuals with experience in genome data analysis, large-scale data visualization, artificial intelligence and machine learning. They will join a cohort developing computational tools and services for the analysis, modeling and management of biomedical data in patient care. Fellows will be responsible for building, modeling and testing algorithms and visualization tools informing the clinical management of patients from large-scale clinical laboratory datasets and patient electronic medical records. Competitive applicants will include individuals from varied training backgrounds, such as medicine, biosciences, engineering and computer science. Fellows with strong computational, analytical and statistical training and bioinformatic and machine learning backgrounds are preferred. A PhD or MD is required.

Committee Structure: Fellows will be assigned to a data science mentor and a clinical mentor that they will interact with on a daily basis. Fellows will rotate with their clinical mentors to understand operational clinical workflows, infrastructure, unmet needs and opportunities for innovation. Fellows will then work in collaboration with their data science and clinical mentors to prototype and develop health technology solutions for clinical deployment.

Program Timeline: During the first week of the program, fellows will participate in group discussions with healthcare and data science experts to establish a foundational understanding of concepts, principles and challenges in clinical care. During this time, fellows will learn about common tools and methods for healthcare data analysis, and will begin to brainstorm potential fellowship projects. Fellows will then rotate and embed themselves within the practice of their clinical mentor. The remainder of the year will be spent collectively with their data science and clinical mentors, working on their projects. At the completion of their first fellowship year, fellows may be able to extend their training for another year, pending availability of a specific data science research mentor. Monthly joint seminars with other fellows, and their clinical and data science mentors, will serve as key opportunities to present project ideas and updates, and solicit feedback from the larger group.

Project Examples: Fellows will work with an interdisciplinary group of researchers, data scientists, clinicians and product developers to build tools and services for a learning-based healthcare system, supporting clinicians and improving patient outcomes. The purpose of this collaborative effort is to address gaps in domain knowledge, remove isolated data silos and understand important clinical challenges, fostering effective interdisciplinary partnerships between experimental and quantitative biologists and healthcare providers. Examples of potential projects include:

  • Healthcare Analytics: Mine electronic healthcare record information to develop learning-based clinical decision support tools and improve clinical management and patient outcomes.
  • Genomics: Perform prospective and retrospective analyses of clinical sequencing data to build predictive models of cancer patient mutational signatures and disease relapse.
  • Quantified Self: Development of tools to integrate personal health sensor data from mobile phone applications and wearables with routine laboratory testing values to better track overall health, early disease diagnosis and disease progression.
  • Clinical Trials and Drug Development: Development of analytical models to identify and target specific populations for clinical trials and novel drug development.

 

Application Process

We invite eligible applicants to submit: 1) a single-page letter of intent addressing your interest in the program, your relevant qualifications and training, and relevant prior projects and research; 2)  a CV; and 3) citizenship/residency/visa status to Helio Costa (hcosta@stanford.eduby October 31, 2018. A selected group of applicants will be invited to prepare a more detailed application and interview with faculty members in mid-December 2018. The fellowship will begin September 2019 and run through August 2020.

Current Fellows

Jessica Chen, Ph.D.

Clinical Data Science Interests: Clinical Genomics, Healthcare Analytics, Drug Discovery

Haik Kalantarian, Ph.D.

Clinical Data Science Interests: Ubiquitous Computing, Wearables, Sensors

Nick Haber, Ph.D.

Clinical Data Science Interests: Artificial Intelligence,  Active Learning, Ubiquitous Computing

Riyika Yamashita, M.D., Ph.D.

Interests: Machine/Deep Learning for Biomedical Data (Radiology, Pathology, Genomics), Clinical Decision Making, Causal Inference, Dataset Bias