Interpretable machine learning models for lung cancer screening

Principal Investigator

Name
Cynthia Rudin

Degrees
PhD

Institution
Duke

Position Title
Associate Professor

Email
cynthia@cs.duke.edu

About this CDAS Project

Study
NLST (Learn more about this study)

Project ID
NLST-332

Initial CDAS Request Approval
Aug 2, 2017

Title
Interpretable machine learning models for lung cancer screening

Summary
The goal of my lab is to produce interpretable machine learning models and test them on publicly available data. We hope to provide clinical tools for lung cancer screening using machine learning where the algorithms not only provide predictions but also explanations for how the predictions were made.

Aims

- Create machine learning tools for interpretable modeling
- Test them on publicly available data
- Work with clinicians to develop measures of interpretability specific to lung cancer screening

Collaborators

Amir Tahmasebi - Philips Amir.Tahmasebi@philips.com
Omar Badawi - Philips - omar.badawi@philips.com
Reza Sharifi Sedeh - Philips - reza.sharifi.sedeh@philips.com
Hongyu Yang - MIT - hongyuy@mit.edu