< Back to all Announcements

Opportunity for Duke Students: Data Science Projects in Medicine

Program Summary:

Data science is having a significant impact in medicine and health. It is anticipated that the clinical delivery of care is going to be significantly impacted (“disrupted”) by advances in machine learning and artificial intelligence as applied to medicine.

This program offers Duke students, both undergraduate and graduate, the opportunity to be a part of research teams applying advanced machine learning (deep learning) to important areas of medicine. The students will work in teams, paired with clinicians and professional data scientists, analyzing real data generated in the Duke Health System. Areas of focus include dermatology, ophthalmology, pathology, radiology and cardiology. In each project the focus will be on applying state-of-the-art deep learning technology to image analysis, with the goal of assisting clinicians in making decisions on diagnosis and delivery of care. The students will be mentored by leading Duke faculty involved in data science research, as well as by practicing clinicians. This program is a component of Duke +DS.

Time Commitment:

Students interested in this program must be prepared to devote time during the Fall 2018 and Spring 2019 semesters. Select students may continue this data science research at Duke, during Summer 2019 as paid interns.

During the Fall 2018 semester, students enrolled in this program will be given training in deep learning, with a particular focus on image analysis. The training will be constituted via the online +DS learning modules, and via complementary in-person learning experiences. Additionally, in the Fall 2018 semester the students will be introduced to the team of clinicians, problem statements, and examples of the data. This introductory training will take place in October, and in November the students will be partitioned into teams, with work beginning on implementing the data science projects. The students on a given team will work together, and information/learnings will be shared between teams.

In the Spring 2019 semester, a class will be designated for this program, to which all participating students must signup (to block the time period). During the Spring 2019 semester significant progress is anticipated on the data science, with the students working closely with the data science and medical faculty mentors. It is anticipated that at least a subset of the work product from this program will result in publishable research results.

It is anticipated that a subset of the highest-performing students will be selected to continue the projects/research, via paid summer internships at Duke, during Summer 2019.

Target Student Population:

This program is appropriate for any Duke student (graduate and undergraduate), in any major, who is interested in advanced data science and its application to health/medicine. There will be significant computer programming in Python, so prior experience with such is useful, but not required. This program is appropriate for students who feel comfortable coding, and the details of machine learning and the application to medicine will be taught (no prior machine learning experience is required). This program is particularly appropriate for students interested in data science, and to pre-med students interested in real applications of machine learning to medicine.

Application Process:

Interested students should send a resume, and up to a one-page statement describing why they are interested. Please send these materials via email to plus-datascience@duke.eduThis material must be received no later than 5pm EST on October 1, 2018.