Deep learning algorithm for the detection of critical findings on chest CT images.

Principal Investigator

Name
Maria Mercedes Serra

Degrees
M.D.

Institution
Entelai

Position Title
Medical Director

Email
mechyserra@gmail.com

About this CDAS Project

Study
NLST (Learn more about this study)

Project ID
NLST-670

Initial CDAS Request Approval
May 12, 2020

Title
Deep learning algorithm for the detection of critical findings on chest CT images.

Summary
In the last decade, there has been a significant increase in the number of CT studies performed yearly in relation to the number of trained radiologists. The report of critical findings is often delayed because of extended worklists. The goal of our project is to develop, validate, and test a deep learning algorithm that detects critical findings on chest CT images and helps reorder radiologists' worklists based on these findings.

Aims

The goal of our project is to develop, validate and test a deep learning algorithm that detects critical findings on chest CT images and helps reorder radiologists' worklists based on these findings. Our specific goals are to detect acute findings including pneumothorax, aortic dissection, pneumonia and pulmonary thromboembolism as red flags; and chronic findings including pulmonary nodules, COOP signs, or interstitial disease, among others, as yellow flags.

Collaborators

Mauricio Farez (CEO, Entelai)
Diego Fernandez Slezak (CTO, Entelai)
Joaquin Seia (Data Scientist, Entelai)